This sentence was a bit cute: "Waymo has received our pilot permit allowing for commercial operations at San Francisco International Airport." Yeah, that kind of pilot.
I really had to read through it twice to make sure they were just talking about car taxis picking up travelers, rather than some kind of prototype pilotless commuter helicopter or something.
The hard part of automated driving is dealing with all the ground clutter that planes serenely fly over. If pedestrians could charge out in front of a 777 going 650 mph at 34,000 feet... well... we'd be living in pretty different world! And in that world, flying would be much more difficult. Not just for computers but for humans too.
Flying is obviously much harder than driving, but it's a sort of harder that is generally more amenable to automation, though I still think pilots are a good idea because when it goes wrong it goes wrong much worse.
Flying is almost always easier than driving. landing is hard. Bad weather is hard. But just flying - human pilots have napped many times over the years and it only rarely is an issue. Airplanes with primitive autopilot are very good.
Yeah, a primitive autopilot in a plane just needs an altimeter and compass, but a AoA sensor, speedometer, fuel level sensor, and pitch sensor help to detect unsafe conditions like runaway pitch, stalling, overspeed, low fuel, etc. Each of those sensors is providing a simple 1-dimensional data point. Redundancy is relatively inexpensive.
Automatic lane keeping in a car requires cameras that software needs to then analyze to find the lines in the road in real time. But if you want a "set it and read a book for an hour", then you have to respond to other traffic. No longer just some simple PID controllers, the software now needs to plan and execute based on surrounding traffic.
Taxiing is probably harder to automate than the rest. But you could have pilots on hand to taxi to the runway, and take a shuttle to the other end and hop on a just landed plane to taxi to the gate. Or you could use tugs for ground movement.
I'm not convinced - in a commercial airport taxiways are controlled by a ground control systems, not just pilots looking out the window. If the only airplanes around are also equips with the self taxi system they just report position to the central control and that tells them when to go. There needs to be emergency overrides for when that system fails, or a small plane without it is around, but that can be handled by stopping everyone else in the area until the hazard is gone.
Do you have a source for that? As to my knowledge advanced systems (such as lights on the twy directing you) are only present at very few airports. Recent incidents even happened due to RWY incursion without a ground controller noticing under bad visibility. So we are at a level where your runway is not even protected accordingly, let alone your 50+ taxiways.
There's also all of the service vehicles when you get closer to the gates. The likely damage from an incident during taxiing is much less than during take off or landing, but I think the risk of having an incident is higher and the situation is trickier to manage. And it's super doable to have a pilot come on to manage that, and drop off after the hard part; you couldn't reasonably have pilots do a takeoff and then jetpack over to an arriving plane to do the landing, but it wouldn't be unreasonable for ground moves... similar to canal pilots taking ships through canals.
"The [German pilots'] union said it had carried out a survey of more than 900 pilots in recent weeks, which found that 93% of them admitted to napping during a flight in the past few months."
-The Guardian, "Almost all German pilots admit to napping during flights in union survey"; 2025-09-10
Years as since humans have flown planes stable enough not to need constant attention. On a calm day you don't need autopilot, just set your trims correctly and some airplanes will hold course well enough for a short nap - though of course this is more likely to result in a crash (which likely has happened, though it is hard to guess why a plane crashed beyond pilot error)
Not to mention that almost all civilian planes in the US are required to broadcast a bunch of details that include their coordinates and altitude on a public channel (ADS-B). It's the kind of automated collision avoidance input that you'd probably dream of as a self-driving system engineer. Basically the only thing you'd need to avoid via more complex systems is the odd military traffic, small craft at low altitudes, and birds.
In the abstract yes but in practice the economic (ratio of cost of pilot to pax miles) and safety context of aviation mean fully autonomous flying has to be extremely robust before it has actual utility in industry.
In practice, you're also currently very reliant on infrastructure that is definitely not as solid as you want (eg: ILS and GPS can be interfered with quite nastily).
ILS being under maintenance and unavailable for certain runways is also far from unusual.
Commercial pilots are also extremely good at dealing with edge cases you wouldn't design an autonomous system for no matter how solid the infrastructure, like deciding the Hudson river is a good place to ditch
And their cost relative to other operational costs is so low there isn't even any pushback on regulations regarding there being two of them.
Pilot cost isn't low. The airline industry is very much looking for ways to reduce crews, whether that's going to single pilot operations (long term) or reduced crew operations (short term).
On the happy path, yes. Though I don’t think takeoff is automated yet.
Currently we rely very much on the problem solving abilities of human pilots to deal with troublesome situations. Autopilot will disengage in many scenarios.
Drones (both autonomous and remote piloted) have much higher mishap rates than crewed aircraft. Taking off is "easy" until something goes wrong, like a mechanical failure or runway incursion. It's impossible to anticipate and explicitly code for every possible failure mode, so developing autonomous flight control systems that would be safe enough for commercial passenger flights is extremely challenging.
Category IIIC ILS (full auto-land) does exist but requires special equipment for both the aircraft and airport. Human pilots have to actively monitor the system and take back control if anything goes wrong (which does happen).
Garmin also has the Autonomí auto-land system for certain general aviation aircraft which can attempt to land at the closest suitable airport. But this is only used for single pilot operation in case the pilot becomes incapacitated. It isn't suitable for regular flights.
OK, I've considered that and determined it to be mostly wrong. While drone failure is an acceptable outcome, current technology still doesn't allow drones to be as safe as equivalent crewed aircraft across the full range of flight operations. Maybe in 50 years we'll get there.
Sure current tech does not allow us to safely automate flights. What I wanted to get at is that tech that doesn't need developing does not get built. Looking at drone failures does not tell us the max safety they would reach if we focused on that.
Obviously a droneliner would look very different from the jets that are common today.
OTOH takeoff and landing could in theory be operated by people on the ground, flying simulator style.
I still believe that having an actual pilot inside the plane that care for his own life is not a bad idea vs someone remote feeling a bit disconnected with the reality of a crash.
Remote piloting is how the military operates certain drones like the MQ-1 Predator. The mishap rate is very high relative to crewed aircraft due to network lag and sensor issues. The military is willing to accept some level of equipment loss in order to accomplish their mission but this would never be allowed for commercial airliners.
The pilot’s self-preservation instincts aren’t the most important reason to have them onboard. It’s that any loss of communication between the ground and the airplane at any point during either procedure would turn it into an uncontrolled cruise missile.
I am not sure why you were down voted. The original meaning of the word pilot is someone who comes aboard a ship for "the last mile" - getting in and out of the harbor and what you are talking about is kinda like that - a person associated with the airport rather than airplane to guide the planes in and out - perhaps using more reliable local communication technology vs what is used to control drones half way around the world.
I have no idea if that works but I thought you were making a good contribution to the conversation by proposing a potential solution to the exact problem everyone is talking about.
I'm not actually sure how hard landing is. Most airports that support autonomous landings do it by having ILS antennae that guide the airplane to within tens of feet of the runway, at which point the airplane switches to radar for altitude.
Automatic landings started in 1964. I think that it seems hard mostly because of how tightly regulated aviation is - modern technology could probably make things a lot better if people were more receptive to the idea of heavy automated aircraft over populated areas.
landing is easy. the hard part is landing with 20mph cross winds and one engine out (or other mechanical failures). we've had auto-land that is 99% reliable for a while now, but you need to get to 6 9s before you have a system safe enough to replace pilots
I think that as long as the autopilot is able to fly in a crosswind or with an engine failure, it can probably land with one. Autopilots are already able to do these things.
I doubt anyone has tested this in depth, but I'm not sure there are too many configurations of airplane these days where a human can safely land it and a computer can't. Maybe if a big chunk of wing or control surfaces were totally gone, but even a human pilot isn't getting 99% reliability in a situation like that.
In any case, I don't think that the first candidates for automation are gonna be passenger flights. It will probably be small cargo planes first - Cessna Caravans and other turboprop aircraft where the cost of paying pilots is roughly similar to the price of fuel.
My layman's understanding is that we've been doing it already for decades with expert system "AI", so likely much easier than navigating streets with other people.
Not a pilot myself but it seems that a large part of the danger with flying is that when something goes wrong you are much more likely to have a high speed crash. Cars don't even usually travel at speeds that planes crash at.
When everything is working correctly, no other pilots have emergencies, and no temporary restrictions are in place, and there are no clouds in the sky. Then yes, it /could/ be easier, but almost always it never actually is.
There's a reason the majority of accidents occur during take off and landing.
It depends a bit on your safety standards. There are already autonomous flying things delivering blood and blowing up oil depots where it doesn't matter so much if stuff goes wrong, but to be an airline pilot you have to know how to deal with a huge range of emergencies and systems packing up.
With a car if the engine fails you just pull over. With an airliner it's not so simple. As a result the training for a pilot is much longer than for a bus driver say.
The problem is actually safety. As automated systems get better, the pilot is left with not much to do, and has to maintain vigilance while being really really bored. It is almost better to have fewer automated systems and give the pilot more things to do during the flight so it is easier to keep them paying attention, or all automated with no human pilot to mess things up.
Depends on the size of the plane, really. One of the reasons a few companies were investing in fully autonomous air taxis is because the math on a small piloted aircraft wasn't realistic for a low enough price point to be competitive.
When shit goes wrong for a car (such as Waymo) you just stop. Now, that's not trivial, but it's also not very difficult, I expect most of it can happen even if the Waymo hardware itself were suddenly destroyed, rolling along is the hard bit so not doing that isn't too difficult. Everybody aboard can just leave when it stops moving.
In contrast when shit goes wrong for a plane we've got a big problem. Just stopping will definitely kill everybody, even from a modest altitude at a very low speed suddenly plummeting to the ground will straight up kill you. So, we want to land, albeit maybe we have to "crash land" destroying the vehicle to perhaps save its occupants.
You can buy (and indeed to some extent you can even retro-fit) emergency auto-land for small planes. Once engaged, or if set to do so automatically upon pilot failure the plane will figure out where it is (using GPS), pick the emergency radio frequency and announce the problem and its intended solution (I am a machine. My human pilot is incapacitated. I intend to fly to X location and land there. I am not listening to you and cannot understand you) and then it will fly to a chosen place and attempt to properly land the aeroplane, broadcast on radio that this airfield is now closed (this aeroplane is parked on the landing strip so you can't use it!) and then switch off.
Maybe the pilot is still alive and human medics can rush them to hospital. Otherwise maybe there are passengers who have been saved. In any case at least the aeroplane is now on the ground where humans can easily take over e.g. moving the plane so the airfield can re-open.
That “just” is doing some heavy lifting! The car still has to deal with all the normal hazards of the road while pulling over, plus the hazards it is itself creating by acting abnormally.
Could also be a big challenge if you have dozens or hundreds of autonomous cars in the area that need manual intervention to get them out (plus the people who get stuck there)
Well if we're being picky, technically the car itself doesn't have to deal with the hazards it has created, rather everyone else does.
The point is you can't just "stop" a plane and wait for someone to figure things out (https://support.google.com/waymo/answer/9449023?hl=en). Whatever the difficulties in dealing with an abnormal situation in a car, it is strictly much more difficult to deal with them in a vehicle constantly fighting the homicidal urge to fall out of the sky.
Ok cool, so a plane merely needs to continue at highway speed, in the face of any difficulty, be it mechanical, electrical, software, weather... Like the movie Speed.
Don’t have a ref but heard that it’s been safe for quite a while but they keep the pilots around due to consumer fear rather than actual improved performance. Curious if anyone can confirm.
No. Airliners can't even take off on their own yet, and are only allowed to auto-land with zero visibility at a few dozen airports when the pilots, plane, and runway are all current/recently checked.
Look up the Airbus ATTOL project's first automated takeoff a few years ago.
Also, there's virtually no automation when it comes to interacting with ATC.
An airplane will take off when it is properly configured and it hits a certain speed. It's simple aerodynamics/physics. Pilots are there to react to failures and unexpected events.
There's a bit more to it since you do need to do last bit of configuration (pull up the nose) just as you hit the target speed. But yeah, automatic take-off is quite a bit easier than automatic rejection of take-off.
Even manually pulling up the nose once you reach Vr isn't necessary if you just trim for a little extra nose-up. It'll eventually get off the ground with just enough speed.
Yeah temperature, wind, altitude, weight, runway slope all matter, and then there needs to be enough spare space for the aircraft to successfully take off even with engine failure in the worst possible moment. Then there's the question of fuel consumption too. Takeoff power isn't typically configured to get the aircraft off the ground as fast as possible, but to minimize fuel consumption, while still leaving enough margin in case of engine failure.
It wouldn't be that hard to fully automate a flight from gate to gate when everything works perfectly. But the various failure modes, human error like airport vehicles entering active runway, all that requires human backup. Self-driving car can just stop to the side of the road and turn on emergency lights if its engine fails, with a plane things get much more complicated.
One of the hardest parts is just getting radio comms right. ICAO phraseology is supposedly standardized but when anything unusual happens then things get messy, especially if there are multiple aircraft involved.
Cars can drive around without needing to talk to other cars or controllers.
And you don't need rudder input or any aileron input because of crosswind, and other bits that falls into "technically correct but not particularly relevant" territory.
It's fun to see/feel planes do stuff "on their own" (eg making them oscillate, or level on their own, or feeling ground effect, or even your own wake on steep turns) but it's not something you'd want to rely on (maybe with the exception of ground effect on short field takeoffs, but I digress).
> Also, there's virtually no automation when it comes to interacting with ATC.
Check out the Cirrus Autoland feature in their aircraft. They are all small personal aircraft, but the tech is pretty cool. Will talk to ATC and fully auto-land for you in the event of an emergency where the pilot is incapacitated.
"Talking to ATC" is a bit of a huge ask. The system basically just hops on 121.5 (and maybe the nearest/local unicom/tower frequency) and start an automated callout with its intentions that it will be doing. It operates on the assumption that all other airspace users will hear the radio calls and stay clear of the emergency aircraft.
If you can design the product and environment to fit automation, then automation can be quick and effective.
The less you can change about the product and environment, then automation run slower and less effectively.
Air liner operations could be automated, but the minimum equipment list would be more stringent, the destination airport would not be able to take any equipment out of service for maintenance, visibility minimums would increase, takeoff and landing operations would require more slack time.
Besides all of that, the owner of the airplane would still want to have some crew on board.
In short, it's not worth it yet.
===
There is also the paradox of automation: Automation generally makes the hard parts harder and the easy parts easier.
The current goal of autonomy for airliners is single-pilot operation more than full autonomy.
It's very cool stuff, technology wise, with potentially significant redesigns of cockpits, etc.
But the main thing is the plane basically needs to be able to operate just about entirely autonomously (especially during critical flight phases) in case the pilot is incapacitated.
In theory, once SPO is solved, autonomy is almost solved.
I'm skeptical that SPO will be allowed for commercial airliners in our lifetimes. Pilot workloads are fairly low during most routine flights. But when an emergency occurs then the workload suddenly gets extremely high, to the extent that even two pilots are sometimes overwhelmed. This isn't a problem that current automation technology can solve. There are an infinite number of possible emergency scenarios and engineers can't possibly code for and test every one.
Cargo flights over oceans and (mostly) unpopulated areas might be a valid use case for SPO. Cargo pilots have always been considered somewhat expendable.
I watched video about incident where plane was really lucky that there was a pilot riding along in the jump seat when engine went out. The pilots were wrestling the plane and the extra guy was able to debug the real problem. Maybe it was figuring out which engine was on fire and shutting it off.
In an automotive setting you can almost always safely decelerate to a full-stop, put on hazards and call it a fail. Good luck trying that in an aircraft over urban areas.
Most carriers have a rule that on clear days you always hand fly the landing.
This is a competence you do not want to lose.
It's also the case that you can have a whole approach setup in your flight computer and at the last minute the controller gives you a runway change. You could drop your head down and start typing a bunch info the FMC but you're generally better off just disabling auto pilot and manually making the adjustment.
But two interesting data points from the Wikipedia article I linked are that the first aircraft certification for ILS Cat III was in 1968, and Cat IIIB in 1975.
And IIRC by the 1980s, autoland was already a pretty common feature.
Yes, but autopilot usually just keeps the plane flying in a straight line at some specified altitude, which have been around since 1912. It isn't full self-flying (although we definitely have drones that can fly themselves already, so that tech already exists).
Auto-landers are not simply classified with autopilots. An autoland system is an advanced function that is part of a modern aircraft's overall autopilot capabilities. A basic autopilot can control an aircraft's attitude and heading, but an autoland system can automatically execute the full landing procedure.
Cool, I wonder if this means they will finally start letting foreign visitors also use the app. I'm an American living abroad now for many years, and I was initially super excited to try Waymo in LA and SF this summer when I visited with my family. Unfortunately they only make the iPhone app available via the US app store, and while I actually have a US credit card that I could have in theory used to make the switch, Apple makes it an absurd pain to change your region as they require you to both a) cancel any existing subscription AND b) wait until they all expire. Most tourists have it worse as they have no option to even switch in theory.
As a European, I can’t help but feel a bit sad that we’re missing out on the driverless side of things. It seems like most of the meaningful deployments are happening in the US (Waymo, Cruise).
I’d really like to see either a Waymo competitor emerge in Europe, or even Waymo themselves operating here. The regulatory environment is obviously more complex, but it’d be great if we didn’t end up years behind on something this transformative.
Cars of any sort, self-driving or otherwise, do not solve traffic any more than Uber does because you need to have enough of them to get everyone to and from work at basically the same time. Trains are the only way to address traffic. Trains are self-driving. Europe already has the better self-driving system. It's just boring because self-driving is much easier when you build the road to support it instead of removing all constraints and adding GPUs, lidar sensors, cameras and an army of fall-back operators in overseas call centers.
If trips that require a car are prohibitively expensive (in money, time or convenience) without owning a car, more people will own a car. Once you own a car, it's often much easier to use it for trips that you would otherwise do without a car.
Reducing the (perceived) need to buy a car, e.g. by making it easy, cheap and reliable to get from A to B using a self driving car service, will reduce the number of people who own a car and thus the number of car trips.
Taxi services can potentially complement public service by filling in the gaps: last-mile connections (home to train station) and backup service late at night when transit runs less frequently or not at all.
There's a risk that robotaxis could become too cheap and people use them for point-to-point transportation because it's faster. This could be mitigated through taxes on robotaxis (with incentives to connect people to transit) and/or car usage in general, or maybe using robo-buses to provide a middle ground between personal convenience and system efficiency.
>Reducing the (perceived) need to buy a car, e.g. by making it easy, cheap and reliable to get from A to B using a self driving car service
But this assumes the need for a car, but cars are one form of transport. A more wholistic look at transportation with be “Reducing the (perceived) need to buy a car, e.g. by making it easy, cheap and reliable to get from A to B.” If you have more services within walking distance, it reduces your need for a car. If there is lots of bike infrastructure, it reduces your need for a car. If there are reliable frequent trains, it reduces your need for a car. If there are reliable frequent bus services and bus lanes to get around traffic, it reduces your need for a car.
On the other hand, if there are more cars then you need, at minimum if we imagine self-driving cars, more road capacity. But realistically more roads and more parking. More space for roads is less space for the actual places people want to go, pushing those things farther apart. Being farther apart reduces the number of places you can get to by walking or biking, which means you are more likely to need a car, which means more cars, which means more roads, which means less space for the actual places people want to go, repeat. Cars are basically the worst option in terms of infrastructure cost, land usage per person, personal cost to use/operate, deaths and injuries, etc.
This assumes the occasional need for a car, because even with the best public transit etc. in the world, there will be cases where not using a car is impractical.
If you need a car on at least a weekly basis, you're probably going to have your own car either way (unless the self driving car services are really good and cheap).
But even if all everyday trips don't require a car, it's very likely there will be some exceptions. And those can make or break this. If getting a car for that occasion requires hours of overhead (e.g. getting to a pick up/drop off point), is sufficiently inflexible (cars not reliably available on short notice), or prohibitively expensive (e.g. per-km charges on car sharing cars that make a couple longer day trips per year more expensive than just getting a cheap car), people who otherwise could do without a car will consider getting one.
OTOH, if the alternative is really good, people who occasionally need a car might use a service rather than owning a car, which means usage-based cost i.e. a much bigger incentive to pick alternatives. If they have been pushed to own a car, the fixed costs are a sunk cost and the marginal cost of taking car can easily be cheaper than public transit.
The main effect of making the car more comfortable, in this case by removing the controls, is to encourage (subsidise) people to spend longer in the car.
So people will be willing to drive further for cheaper rent, or the self driving car might add a couple extra miles to park somewhere cheaper, so overall congestion would get worse.
Can you imagine how much traffic there would be if NYC didn't have the MTA? The principle of induced demand tells us that as long as there are roads they will have roughly constant traffic because people are willing to spend some roughly constant amount of time getting to and from destinations by road each day. More roads speeds up everyone's commute which brings in more drivers, which brings traffic right back to the baseline terribleness.
The question is how shitty it would be if they also had everyone on them who's currently on public transit.
So basically, it is a traffic-free panacea for everyone who chooses to use it. It's not a goal of trains to eliminate traffic for everyone who insists on driving.
The induced demand argument works for trains too. If NYC didn't have MTA (no subway, no LIRR, no MNR) then the population of NYC would probably be 1% of what it currently is. Building more train tracks and having better train services also encourages more people to move to NYC so that these new train services become more utilized.
Neither roads or train tracks solve the traffic problem.
Train density is high enough that you might actually be able to build enough tracks to keep up with demand. Tokyo has just about kept up with growth by building trains, and (unlike cars in NYC) the trains don't have to dominate the city to do that.
Yep, this is a good point. There are appropriate technologies for each situation. It's not a winner-takes-all contest.
For another example, can you imagine trains replacing school buses in a large, rural school district? Sometimes (not always), buses are better than trains.
Any one part would have the about same amount of traffic it does now. It would just sprawl out bigger across adjacent counties and the highest density parts would be lower density.
This is what bikes and busses are for, or just walking because the metro system is comprehensive enough you are at most four blocks away from a station.
You could maybe have something like Zermatt Switzerland which is car free but you can get around in human driven golf cart like taxis. It's pretty pleasant but expensive. If the carts were self driving it could be cheaper.
Zermatt is fundamentally a pedestrian town. There are a limited number of permits for electric vehicles available for companies that have an objective need for a vehicle. That limited availability makes the electric taxis expensive.
The total number of permits seems to be around 500 in a town of 5k permanent residents. And the population grows to 30k or 40k during the peak tourist season.
Yeah that approach can't be scaled to cities. Folks go there to chill or do alpinism, not live their lives and work. Otherwise those narrow steep streets would have very quickly rush hours and traffic jams, its really not a place designed for any traffic apart from walking.
One day, cheap automated electric self driving taxis will cover cities, thats unavoidable I think, but we are not there yet.
Cities are experimenting with traffic free areas like Barcelona's superblocks. You could imagine something like that but with cheap automated electric self driving taxis added. I agree we are not there - Waymo basically just substitute normal taxis.
> Trains are all very well but they've been around nearly 200 years and have yet to bring on a traffic free utopia.
Cars will always have a purpose. But if you go to somewhere like The Netherlands, they are much less relied upon - it's more about delivery vans than getting individuals to places.
I think self-driving cars can still be beneficial even if they don’t help with traffic problems. They shouldn’t require so much parking in desirable areas (a separate problem cars cause), for example, and they could have a big impact on the lives of some disabled people.
Trains will fairly unreliably take you from one place that is not your home, to another place, which is not where you want to go, at a time that is probably not exactly when you wanted to arrive. Freedom of movement is incredibly important, and trains are very rigid in this aspect.
Well That’s certainly not been my experience when visiting Europe. In fact, it many cases it’s been the opposite - having a car would have been restrictive in any major city and a source of friction.
Well to the extent it draws people from public transit, yes because traffic makes being a pedestrian more unpleasant and waymos still are traffic. And increased traffic adds friction to crossing streets and they park obnoxiously, among other things.
So yes, they would be obnoxious at any significant quantity and also not really help with getting across the city since transit is pretty good
Well yes if we’re arbitrary limiting our choice to car based transportation that makes sense for mild climate cities. But why are we insisting on cars being the backbone?
No limits. Each option should be evaluated on its merits.
My contention is that in US cities the high cost of existing rail makes it uncompetitive for most uses, and there is no justification for building new rail.
Maybe not a greenfield project, but rail lines like the NEC could benefit a lot from relatively cheap fixes: removing sharp curves, improving scheduling operations, etc. We just need to get the flywheel going on this in the US
I like Waymo a lot, but the USA desperately needs both transport modes. Don’t think it’s an either/or.
First, I assume that "NEC" means North East Corridor which has a "high speed" train on Boston-NYC-WashingtonDC. Second, "relatively cheap fixes: removing sharp curves": You lost me here. That train must be about 20 years old now. If this was so cheap and easy, why not already done?
Hell if I know why it hasn’t already been done. All I’m saying is that the route slows down because of some sharper turns in some areas, and fixing it would be easier than making completely new lines/stations. I’m sure it would be much more expensive than similar projects elsewhere in the world
Cars driving at high speed over normal asphalt also generate a lot of tire noies and particulate pollution, even if they are electric cars. I found this video pretty interesting - some cities are experimenting with different road surfaces to reduce noise
This is quite the "I have never lived anywhere else other than North America" take.
Rail and other public transport in pretty much everywhere in the world are designed to serve commute first, tourist stuff second or third.
Public transport isn't just having some trains, or having only trains between major cities. It is designing whole commute routes from various urban and suburban areas to workplace. There needs to be regional and suburban links that arrive to metro and tram stations. Metro and tram have to operate very frequently to handle commuters. The frequency of the trains should adapt to the commuters in the morning and evening. They need to be convenient, clean and safe too.
Cities around the world are also much better balanced than NA ones. The workplaces and living areas are almost always mixed rather than having a "downtown" area where every office worker travels to. My area has many buildings with a supermarket, apartments and small offices in the same building. There are two car factories in the city next to one of the biggest urban parks.
I'm a European who has emigrated to the US, and knows both sides pretty well.
I agree that European trains work very well for commuting to and from the center of big cities. That's where the jobs and tourist attractions are.
But to go between arbitrary places A and B is usually quite painful. Often the best way is to go to the center, and then from there to your destination.
When I moved to the US and got a car, it was an unreal feeling! I could quickly travel anywhere at anytime!! Practically it felt like my comfortable travel radius increased from 10km to 50km.
> Trains are great when going to tourist attractions, especially in the center of old cities. When you live and work in a city, they're much less practical.
This is the most "tell me you live in America without telling me you live in America" thing I've seen in a long time...
America basically the only place in the world where in its cities, trains and other public transport aren't a major part of people's lives. In other places (Seoul, Tokyo, many European cities, etc.), even people who own a car will sometimes commute via train due to the convenience.
In effective countries trains run frequently enough that you don't need to consult a schedule and are less prone to unexpected delays than cars. Yes, they can't provide door-to-door service; like it or not, everyone travelling door-to-door in their private mobile living room during the rush hour is impossible if you want cities dense enough to be liveable.
Try a bicycle or a stroll instead of embracing the WALL-E.
If you feel that way about transit you may not have tried a good transit option like Hong Kong MTR with 90 second headways and travel from and to substantially everywhere you want to be.
Well for my commute the trains are every 30 mins or so - pretty convenient times and a short walk from the office. The ticket is cheap, much cheaper than a days parking and during the trip I get to sit, look at the view and sip a coffee. The train is way more relaxing than the equivalent drive - which due to traffic levels at rush hour would probably take twice as long (at least) and be extremely unpredictable.
So when I have the option I'd rather take the train - of course I also drive a lot of places.
This is based on my personal experience, I used to ride trains for travel a lot. I grew up in Europe and lived there for 31 years so this is not based on ignorance.
ok? Your personal experience is not the entire truth and never will be. Japanese trains are on time. Swiss trains are on time. That's not based on ignorance either.
You haven't been in Washington DC traffic, I take it. It isn't like it is a choice between breezing to your destination unimpeded in your car to arrive reliably on time, and taking the variable, unreliable Metro.
Buddy the tube seldom fails for that reason either. Plus some self-driving sauce would reduce their hours to 0. Certain lines in London like the DLR are already driverless (Grade of Automation 3). Most of the other lines are GoA2.
I live in Switzerland and commute mainly via public transport. We're very privileged here.
Because of decades to centuries of investment, holistic planning and expertise, we have one of the best networks in terms of quality, punctuality and density.
It's a plant the trees for future generations kind of deal, especially in Switzerland, because large, "flashy" projects are rare compared to to the more continuous and steady improvements, due to how funding and planning are set up.
Trains are one part of a larger transportation system. And they are very good at what they do. But you also need metros, trams, buses and so on.
And with that you can build a system where most places, including 50 people mountain villages are well connected.
I can go from Genf to a tiny village deep in the Eastern mountains with 4-6h. I can make that journey with no planning ahead what so ever.
Cars are actually restrictive. What if you want to have a drink? What if you are in a place that is different from your car? What if you are old or disabled? What if you are a tourist? What if you are not allowed to drive because of a traffic violation? And there are also these people called 'kids'. When I was 15 I went from Switzerland to Czechia with the train, no problem.
True freedom is to have a good public transit and potentially car as an option.
Oh the horror, you might have to walk a couple of minutes (probably less time wasted than circling around to find a parking spot, and then walking from it to your destination).
> at a time that is probably not exactly when you wanted to arrive
Yeah, no. Trains in properly developed networks are extremely frequent. At the off-peakest of off-peak (Sunday late evening), the RER near me is every 15 minutes. During peak hours it's every 5 mins.
I disagree that self-driving won't reduce traffic, at least from the perspective a Virginia resident. Commuting into D.C. is in theory very quick, except for when there are crashes. Crashes double the commute time, and there's _always_ a crash. This is pretty much the only source of traffic in my area. I think the primary benefit of self-driving would be lowering the crash rate, and as a side effect traffic.
This gets brought up a lot but I think it's missing some key points.
1) Being driven around is the best transportation mode for most of the US. It's very comfortable, private, fast, and point-to-point. It stops working well at very high density, but that level of density is only seen in a few places in the US. I'd like more people to live in dense areas but for the foreseeable future self-driving vehicles are going to be the best solution for most trips in the US.
2) At very high densities it's true that cars can move fewer people per hour per 10-foot lane than other modes and so you run into congestion. But that's measured with the current vehicle fleet and human drivers. With high autonomous vehicle penetration you could implement congestion pricing that encourages high throughput vehicle design. That means private vehicles that are much much smaller (think Isetta-like design) that can follow at very short distances. Along with the elimination of on-street parking we could see a many-fold increase in road throughput.
3) At even higher density levels the same congestion pricing mechanism would encourage people to use microbuses that would operate similarly to Uber Pool. Compared to today's busses they would have equal or greater throughput, be point-to-point or nearly point-to-point, dynamically routed, cheaper to operate and faster.
4) At the very highest density levels it's true that nothing can match the throughput of the subway. As others have mentioned, AVs are a great way to connect people to the subway. Many trips intersect with the highest density urban core for only a fraction of the journey. More people would take the subway if they knew they could get to and from the stations easily and quickly. AVs let you mix-and-match transport modes more easily.
Cities should start engaging with vehicle manufacturers to start getting these high density vehicle designs worked on and figure out the congestion pricing mechanism to properly incentive their rollout.
As with many "tech innovations" in the transportation space, this rapidly turns into reinventing the bus. #3 in particular is just "the bus, but more frequent" which you can do by simply increasing bus frequency dramatically, which most American cities should already be doing but don't, because of their budget priorities and the stigma of buses as something for poor people.
This rings less like some missing key points, and more like an entire, comprehensive traffic strategy. I'm not really sure what the point is meant to elaborate on. Maybe something like "Self driving cars in themselves wouldn't solve traffic, but well designed, purpose-built AV's combined with surge pricing and (when necessary, depending on the location and journey) trains/subways could do it." Did I understand you correctly?
As mostly a cyclist (I drive roughly 10% of my transport, the rest is biking and transit), my experience with self-driving cars is that I feel much safer riding in front of them. They're less likely to pass dangerously close to me to drive past me, they're less likely to tailgate me, they're also less likely to just drive me into the door zone, sidewalk, or a parked car. I'm a very confident cyclist but I suspect newer, more skittish cyclists would agree.
If you can restrict certain roads to autonomous cars (or heavily limit the number of non-autonomous cars) then you don't need to build as much bicycle infrastructure (a buffered lane is probably all you need, as opposed to bollards or true grade separation) and I can guarantee you more folks will feel comfortable riding bikes. This is aside from how frequently human-driven cars end up colliding with, damaging, or blocking non-grade-separated forms of transit.
> It's just boring because self-driving is much easier when you build the road to support it instead of removing all constraints and adding GPUs, lidar sensors, cameras and an army of fall-back operators in overseas call centers.
I do bike advocacy so this kind of rhetorical gotcha can make me feel good and hit the upvote button but in reality city councils and other elected officials are mostly people skeptical of the benefits of bicycling, worried that buses/trains would place too high a tax burden on their constituents, or deep down convinced in their lizard brain that Americans are too carpilled to ever do anything else. If you can change this by running for your local council, do it!
Don't get me wrong, we need more bike infrastructure and we needed it yesterday. But anything helps. I'd love to see certain corridors of SF be restricted to transit, autonomous vehicle, and cyclist usage only. Market is already only for transit and cyclists so there's precedent.
Bicycles are another way to address traffic, because they take up so little room and can be essentially free and often more convenient for shorter trips. Of course that means you have to have bicycle infrastructure where you don't have to run serious risks to your life every 3-5 minutes during your journey.
Cars as a shared service (shuttles, Uber, Waymo) absolutely solve traffic compared to personal vehicles. Shared cars have much higher utilization and require a lot less space.
I agree that trains are a fantastic way to move large groups of people, but a world with more shared cars (which may be brought about faster with Waymo) is a good thing for most cities.
Do I really care about traffic if I’m not the one driving in it? I guess if you’re looking at highly disproportionate delays but I really wouldn’t care about traffic otherwise.
> Do I really care about traffic if I’m not the one driving in it?
As someone who took the N across San Francisco every day for 5+ years: Yes, you would. Imagine a 5 mile journey taking 50 minutes. Even if you can nap or listen to a podcast, it's still a waste of time.
Trains still don't solve last mile transport for most people (even in places with robust transit systems)
Self driving cars might not solve traffic problems but they could greatly reduce them. Problems like traffic waves and gridlock go away when all cars are driving themselves.
The last mile is a solved problem. Most people can walk (and many of those who can't would need human assistance anyway). And then there are bikes, electric scooters, and other light vehicles that use space much more efficiently than a car.
Self-driving cars may help with the actual weakness of transit, which is the long tail of trips. Trips on routes with too few passengers to justify good transit service, and with the trips too long for the last-mile solutions.
Walking a mile with groceries or a baby is common. People in less car-oriented neighborhoods typically do quick visits to a grocery store when it's convenient for them several times a week, rather getting a week's haul of groceries in a single visit.
With furniture, you usually pay for delivery. Especially because the furniture store probably doesn't have the items you bought on site anyway.
When there are grocery stores within easy walking distance, people tend to grab the next few days’ worth of groceries. When everything is inconveniently far apart, people drive forever away to get giant stocks of things from Costco to haul back to their house.
If people aren’t spending $12k a year[1] to own a car, paying $50-150 to have a large piece of furniture delivered isn’t a big deal.
When there is well maintained, pedestrian friendly infrastructure, instead of a tiny uneven sidewalk inches away from 45 mph traffic, pushing your baby stroller home is not an issue.
The last mile problem is only a problem because of poor layout. Build homes and work near transit nodes (instead of in the middle of nowhere) and there isn't a problem in the first place.
> Problems like traffic waves and gridlock go away when all cars are driving themselves.
How would that make those problems go away? It could probably slightly alleviate them in marginal cases, but any given road has a finite throughput limitation, and once it is reached, it wouldn't matter even if every robo-driver were perfectly synchronized.
> Trains still don't solve last mile transport for most people
This has not been my experience since moving to Manhattan last January. Subways, alone, close the gap between regional rail and most destinations astoundingly well. I haven't yet needed to use a bus (but they seem abundant, too), and I haven't even thought of taking a taxi yet.
Here, robust transit has solved the last mile problem for most people.
Here, robust transit has solved the last mile problem for most people.
There are huge gaps in subway coverage in New York. Manhattan, especially Lower Manhattan, is the exception here. Go to the outer reaches of Queens and see where the subway gets you. Try to go between (or sometimes within) boroughs.
Sure, in areas without robust transit, transit is a problem. But I'm responding to RandallBrown's assertion that there's a persistent last mile issue in areas with robust transit. There's not. Manhattan is evidence that robust transit solves the last mile problem for most people.
Weasel words are doing a lot of heavy lifting here. There is a persistent last mile issue even in NYC, even in Manhattan. You're right that in Manhattan most people can use the subway as a last mile solution. However that map hasn't changed much in quite a while. The subway deserts that exist (in Manhattan and the other boroughs) aren't going away anytime soon because building new subways is eyewateringly expensive.
The inflexibility means that even when the subway is a viable last mile solution it may not be the appropriate one. For instance I had to go from Ridgewood to JFK a few years back. I was maybe a five minute walk from the subway. But were I to take the subway from one end of Queens to the other I would've had to go all the way to Midtown and transfer to LIRR.
Hell I've generally had to rely on buses for last mile connectivity even in London which certainly doesn't suffer from a lack of subway service.
self driving cars will increase traffic as they remove barriers that prevent people that cannot drive from using cars, thus increasing the amount of cars on the road.
They replace taxis and potentially postal and trucking applications in future.
It’s certainly not a replacement for mass transit. US is sparsely populated compared to Europe and mass transit don’t work as well in the suburbia. That said, I do see many transit oriented development in SF Bay Area where high density buildings are being built near transit stations.
Ideally: there's a train close enough to walk, or a bus or tram that's nearby that runs frequently, is clean, and doesn't get stuck in traffic because there's not much car traffic.
Slightly more realistic: enough people can and do cycle or walk to the train that pressure is relieved on the roads for those who cannot cycle or walk.
> And how do you get to the train when it's too far to walk and you're not a cyclist?
You get the bus, or you cycle, which is a life skill any able-bodied adult should have, not limited to cyclists. Of course not everyone is capable of cycling, but not everyone is capable of driving either.
The elephant in the room is rideshare commuting is for extremely rich people. Who else can afford the probably $75+ a day it costs on a two way commute?
Yes, driverless does not solve any real problem. When I come from work, I still have to sit in a car. Yes, I can work instead of drive, but that's only in theory because in practice the G-forces won't allow me to.
A robot cook, however, __would__ solve a practical problem for me.
Anyway, this whole approach is not even solving first-world problems (many families struggle to pay for a car), but it's solving the upper-1% of first-world problems, maybe. Except those people can afford to pay drivers who are now out of a job. So yes, what is this even solving??
Trains are not panacea some people here keep thinking they are. You would need to have train stops every few hundred meters changing it into some city subway or tram, interconnected with dense and fast local public transport.
I live in Switzerland, the place for trains, efficiency and its small and dense, an ideal situation right. Tons of people use trains every day, tons of people also bike for closer distances in good warmish weather but still highways are chock full and getting fuller every year. Public transport for out-of-city commuters is simply slower, often much slower.
This morning I was considering taking a motorbike to a train station that is 5km away, then 40 mins trains and 10 minute walk to work. I took the car instead for a change, I was faster despite having to cross the very center of bottlenecked and car-hostile big city (Geneva) in top rush hour. 65 mins door-to-door via public transport vs 45 in car. That's one way, meaning 40 minutes of my private life daily saved that I can spend ie with my kids and not staring in the phone or out of window.
Normally I take the motorbike if weather permits, if not I take the public bus to the train, adding additional 15 minutes each way. That sucks pretty badly. I doubt other countries have this figured out better, and not everybody can or wants to live in city centers, especially when raising small kids. We did it for 10 years, had a work commute of 5mins via escooter, but I rather have current commute and live and raise kids in small commune next to wild forest and vineyards than that.
All above is usually much worse in many parts of US.
For the consumer, maybe not, other than a delay of some years.
In terms of having the industry? Absolutely. How many other areas of "tech" has Europe basically punted on and ceded to Americans? Currently there's some gnashing of teeth across the pond for how there's no real European equivalent to the big US cloud providers (AWS, Azure, GCP).
There doesn't have to be an equivalent of everything, I wouldn't want to use US cloud because of price and governance. At most I use the "cloudy" services and rent "capacity" from a European provider, companies are fleeing the cloud. They're done subsidizing Amazon deliveries.
MobilEye and Mercedes works on self-driving, so does BMW. It's probably not Waymo quality, but just because there aren't cars on the (wide and car friendly) roads doesn't mean nothing is happening.
Meanwhile Europe has solid infrastructure for electricity (esp France), ASML has no competition, Carl Zeizz is world leading in optics, there's probably a Leica LIDAR in the Waymo cars... I mean while we're throwing pies and bringing up other markets..
My old boss was working on a project with Leica where he was working with some partner on self-driving industrial machines, they we're using Leica gear for collosion avoidance and such.
Europe doesn't need self-driving cars, we have alternative modes of transportation. Where it's needed (mines and industry) it's already there. And whatever modern car you're driving here has ADAS which helps make driving comforable.
Yes, it's fine to give up the lead in any one subsector, but Europe is so far behind in tech industries in general. It's not just cloud services or self driving cars, look at SpaceX and Starlink: Europe has no equivalent to either, and is many years from gaining one (I'm aware of some plans, but they're far away from being able to actually launch, and some are dubious besides).
Both major smartphones OSes? Run by American companies. Major desktop OSes? Two by American companies, one originally started by a Finn, who still manages it...and he moved to Oregon.
But you don't have to take an American's word for it, just read Mario Draghi's report. The man loves Europe, deeply understands the European economy, and has a whole lot to say: https://en.m.wikipedia.org/wiki/Draghi_report
So any b2c thing where you're going to abuse your customers is American, what an achievement!
There's no denying America has done good in some industries, but when it comes at the cost of societies weak I can't help but think it doesn't matter.
SpaceX and Starklink aren't very important to me, I don't know who they're important for except Ukraine, boat and RV owners.
The report says we must invest in electricity infrastructure, well sure so the dude compares against China and USA at the same time? Crumbling infrastructure is the definition of USA 2025.
The cope is American Exceptionalism, we're doing just fine even though we're fighting a unprofitable proxy-war and missed all those b2c investments to leech off humanity.
There's no desktop OS from Finland, that's a kernel and yes he's now American as you guys usually were better at finding ways to turn good into profit.
> The cope is American Exceptionalism, we're doing just fine even though we're fighting a unprofitable proxy-war and missed all those b2c investments to leech off humanity.
We are not doing just fine. We have low economic growth. We are unable to beat off Russia's attack on Ukraine without American help. Germany has crumbling infrastructure just as much or more than America. We have not contributed seriously to any important innovation wave since before... 2000? The invention of the PC?
Economic growth is a measure of how much goods and services are available to everyone. If that isn't improving, that means your quality of life is lower, ceteris paribus. It means you don't produce enough energy on your own are dependent on Russian gas. It means you don't have enough surplus to sustain a military.
I assumed the parent was referring to "GDP growth" which doesn't matter when inflation eats it all and new coins go to megacorps rather than back into society, European standards of living has been consistently improving, especially for the poorer nations.
I can't defend Germany for refusing nuclear in favor of Russian gas, but at the time it seemed to some like a good idea to strengthen relationships through trade and encourage democratization.
It's a damn shame that we're buying Russian gas, it's hilarious that I keep hearing about this from Americans but not Ukrainians.
USA is huge.
This is happening in a small part of the USA in a very limited fashion.
It's not like the USA has driverless cars everywhere, 99.9% of the population never saw one.
I'd guess Waymo covers 5% now. San Francisco, San Jose, Los Angeles, Austin, and Phoenix are ~10% of US population. Waymo service areas don't cover all of those cities.
Considering tourism and people living just outside service areas who see them but don't get to use them (which includes me sadly) I would not be surprised if 10% of population had seen at least one.
> San Francisco, San Jose, Los Angeles, Austin, and Phoenix are ~10% of US population.
Surely you're describing metro areas? There's no way those five cities add up to 34 million people within city limits, given that none of them have 6 million people.
- SF doesn't cover East Bay (two thirds of the MSA by population).
- Silicon Valley doesn't cover San Jose, and barely reaches into Sunnyvale (basically just covering the Google Moffett Park office buildings).
- The Phoenix area is missing most of the densest parts of Phoenix itself, as well as anything north / west of the city.
- Los Angeles doesn't even come close to covering the city, much less the rest of LA County or any of Orange County. (Maybe 2-3 million out of 13, from just eyeballing the region.)
On Uber (https://support.google.com/waymo/answer/16011725?hl=en) there's also Atlanta (which looks like it actually has very nice coverage, other than the western half of the city) and Austin (again focused on downtown / commercial districts) which help drive up the numbers.
The population that's had opportunity to see Waymo in the wild is probably higher because they're testing in quite a few cities now (a sibling commenter mentions NYC, for instance).
>pilots of self-driving taxi- and bus-like services will be brought forward by a year to spring 2026, attracting investment and making the UK one of the world leaders in this technology
I'm wondering how self-driving cars will solve the priority problem of narrow streets of UK towns where drivers need to let each other pass all the time.
I've wondered that myself. It seems quite challenging for human drivers at times. Around Ladbroke Grove you quite often get some complicated jam with two busses and about ten cars stuck.
Yeah I'll believe this when I see it. Most UK roads are significantly harder to drive on than anything in the US. That's why they always test these things in Milton Keynes.
Also a lot of UK driving requires communication with other drivers (letting people out, etc.) in a way that US roads don't. I'm not sure how driverless cars can handle that.
I really wish we could get them, because they're great. But I'd say we're talking 10 years behind the US simply because of the extra engineering challenge.
What do you think degrowth and decline means? Vibes and essays?
It's not just driverless cars either - delivery drones (e.g. in China), a lot of health tech (as they have more check-ups in the USA), Starlink, Neuralink, a space programme, etc.
> I’d really like to see either a Waymo competitor emerge in Europe, or even Waymo themselves operating here
I think you’ll see American and Chinese self-driving kit in Europe once it matures. It’s just easier to iterate at home, so while the technology advances that’s where it will be.
Maybe there just not enough interest? After all there is good public transportation (especially rail), increasing biking habits and just loving the driving experience.
I feel the opposite. Self driving cars seem like a meme because driving is fun and trains are better. If either of those premises is not true in your geolocation then self driving is not the solution either.
I don’t think wages in europe are high enough to sustain this model business very well. When you track waymo deployment its in placed where plenty of high income price insensitive people are to be found.
European cities have lots of taxis. Same with Asian cities. They will obviously have AVs in the future. I'm not sure why you think they should be mutually exclusive with transit.
Many American cities don't have the population density to make metros and trams economically viable. And those few cities that do have comparable density (New York, Chicago, namely) do have metros.
Public infrastructure has high overhead costs, and low population density means there isn't enough ridership to make it viable.
The problem is when cities treat car infrastructure as absolutely mandatory, and all other transport infrastructure (pedestrian, cycle, bus, tram, train) as optional. When you say that everyone has to be able to get everywhere by car all at the same time, you have to build more roads and parking (at minimum more roads using taxis, self-driving), more roads spread everything farther apart, which means more distance per trip, which means more cars on the road, which means more roads, which means everything is spread farther apart, rinse, repeat.
American cities low density is a direct result of designing for car infrastructure over all else. And car infrastructure is far more expensive than other transportation, and since increased car infrastructure lowers density, it directly makes all other transportation more expensive and less viable.
Since cars are the most dangerous form of transport, for other drives but more so for cycles and pedestrians, it makes it less feasible to use them for your first-last mile. Then you add in that, as the roads grow and distances multiply, speeds are increased to attempt to compensate, multiplying the danger to anyone not in a car.
People in the thread are asking why people ride Waymo to SFO, which is well outside San Francisco proper. Thus, the whole peninsula's density is what's relevant.
The US does not have many metro areas with population densities above 3,000/km2. And those that do, like Washington D.C, NYC, Boston, Chicago, do have metro systems.
American public transit construction costs are now ridiculous in terms of both money and political capital. Even somewhere as sprawled as San Jose now requires well over 1b/mi to build a subway under; BART could've acquired an entire autonomous driving company for the cost of the Silicon Valley extension.
…What? What sort of terminally online strawman would be spending his free time “virtue-signaling with Europe” to some anonymous bozos on a tech forum? What a dull and intellectually uncurious reply.
I think self-driving cars may eventually become common in areas where cars are currently common. I think public transit will continue to dominate in parts of the world where it currently dominates, because it is simply a superior user experience for the majority of people when the government cares to invest in it. (Not to mention far cheaper and more egalitarian.)
I am conveying my lived experience in most European cities I've been to.
A superior user experience is going exactly from where I am to where I want to be safely, quickly, and affordably. Self-driving cars are looking really good for those criteria.
$20+ per ride is affordable? Waiting 10m+ for your ride and slowly sifting through traffic is quick?
In London, Paris, or St. Petersburg, I pay a few bucks to hop on a train that runs every few minutes and rapidly end up across town, roughly in the area I need to be. It's literally the cheapest and fastest way to get from point A to point B, not to mention tested at scale and thoroughly battle-hardened over the course of a century.
Not every city has this privilege, of course, but surface trams are 80% of the way there, especially if they have right-of-way. And they don't make pedestrians' lives a living hell: https://www.youtube.com/watch?v=bNTg9EX7MLw
you're shouting at a wall here. This site is absolutely infested with US techbros who believe that the solution to any problem is zero regulation and more computers.
This is why self-driving cars appeal to this crowd. You and i seem to be from a world where public infrastructure like clean, affordable transit is the goal. This raises the floor for everyone. Many here would rather think solely of their own comfort, which is fine, but despite repeatedly being told that they are short-sighted, they refuse to change.
The core for a good experience is a good structure.
In many regions of the U.S. people live too far apart, shops and businesses are zoned apart into wide spread business areas. Public transport won't provide a good experience.
In a notable part of European cities people live in denser quarters, where a "third place" is reachable in walking distance, some degree of shipping, doctor visits, work are close by. There public transport can fill the gaps for the remaining trips in an (space) efficient way. Self driving cars however would clog the area.
Adapting US settlement structure to allow public transport won't happen. However a self-driving car can turn the dial for individuals to move out of the urban European area into more rural areas. Question is how big that group is.
0.1%? You think so? Sorry you're wrong. Suburban population is the largest demographic in North America.
And Home Depot says otherwise. They have reported record profits year over year for the past two decades. Just because you don't use sod in your condo doesn't mean suburbanites don't need it for their homes.
Saying we shouldn't design around that use case isn't being a jerk.
And the exact number wasn't the point. The percent of consumer vehicles on the road that are carrying a significant payload to/from home is pretty small. Especially areas where transit even halfway makes sense. What's your best estimate?
Where I live, the percent of cars that carry a load that would be unwieldy to manage on public transit at least once per month has to be at least 50% and probably closer to 90%.
From Costco trips to babies to wagons, strollers, wheel chairs, hardware stores, bigger box purchases like a TV, out of town trips to visit friends, pet grooming, airport trips with luggage, it's hard for me to imagine a life without a car.
I know you can just say that I'm a product of my circumstances and culture and you don't need a car for any of that, or there are other ways to accomplish my goals, but I could say the same back to you. And the arrow of time seems to point to people everywhere moving in the direction of wanting personal mobility whether horses, bikes, or cars.
It's not all or nothing, but it seems to make sense to me to build around cars as a first class concern, in addition to other forms of transit. Some places in Europe obviously can't, for historical reasons, but I don't see that as a benefit per se, so much as something to have to work around.
Edit: I should add, I did live car free in Boston for 10 years and loved it and didn't really perceive any shortcomings at the time, and even hated having to buy a car when I moved. But now in my 40s with two young kids and a house and an elderly mother, it's an entirely different situation and I can't see how it would work. I would suggest if you're totally anti-car but only in your 20s or early 30s, your opinion might change as your circumstances do. I also lived for a year without a car in Singapore and that was tolerable in a way that wouldn't have been in most places, since it has some of the best public transit in the world, but even there cars are considered luxuries and it would have made things a little easier.
No no no, not the percentage of cars that sometimes carry a load, the percentage of cars on the road that are currently carrying such a load.
If you do that once a week, then you can use transit the other 90% of the time. If people use transit 90% of the time, then we can build smaller roads and de-prioritize cars. That's the argument here, that transit can dominate in co-existence with self-driving cars, not that we'd need to get rid of cars. And especially in the context of waymo there's no effect of "I'm already paying a ton of money to own and insure a car, I might as well use it every trip".
(And again, this is in moderately dense areas where transit works and you actually care about how many cars are on the road to begin with. And it doesn't have to be 90% in particular.)
Good. Cars ruin walkable cities, and the last-mile problem can be solved in other ways.
And it's not just the EU. I'm sure that e.g. China and Japan will continue to invest in their excellent public transit infrastructure even when there are more self-driving cars on the road.
Much of Japan's transit infrastructure is private. There's nothing special about transit that means the government has to own it; being a government, it can regulate things without owning them.
Americans have this idea that transit is for poor people, which translates to "it's not important for transit to make money", which translates to "we need to make it illegal for transit to possibly try to make money", so there aren't even vending machines at the platforms. Whereas in Asia they do profitable land development at the transit stations.
> Much of Japan's transit infrastructure is private. There's nothing special about transit that means the government has to own it; being a government, it can regulate things without owning them.
Japan's private transit infrastructure is only private in high-very high density environments (inner-city) and subsidized in low-density environments (rural, cross-country). Ultimately private group transit requires population density above a certain threshold to be viable.
In Switzerland the Airport has 28 trains per hour that connect it directly to almost every part of the country. In addition to that there is a tram line and many bus lanes.
But I guess in SF they can take a taxi that might be a little cheaper because the company operating it is fine with losing 100s of millions a year.
I think navigating European roads is a massive step more difficult than US cities. They've got wide lanes and a really strict grid layout generally. At least in the European cities I'm familiar with we have much narrower lanes, residential areas with parking turning 2 lanes into 1, old towns, and layouts that are completely unpredictable. Maybe I'm wrong but I think this is the bigger hurdle than regulation.
One thing you are missing out on: mandatory loud (97 to 112 db) 1000 Hz audible beep when the vehicle reversing, oh so slowly, such as at the recharging station. Also, constant shop vac five horsepower vacuum cleaner sound. BEEP BEEP BEEP BEEP BEEP. VROOOOOOOOOOOOOOOOOOOOOOOOOOOOM
Oh wait, you thought those would be in the middle of nowhere? Nope.
It is not mandatory that backup alarms be 97 to 112 db. They only need to be "above the surrounding noise level". The loud beeping alarms were installed on most vehicles because most of them operated at loud constructions sites, so needed to be louder than that. it was easier to just buy the loud model to CYA, even if it was a delivery truck. They also don't need to 1000 Hz or to actually beep. White noise backup alarms are allowed, and in use in many delivery trucks now, and make a sound attenuated above 4000 Hz, which is much more localized and dissipates over much shorter distances. Waymo could absolutely have installed 85dB white noise alarms but chose to install 112dB beeping alarms. This is not a regulation problem.
These backup warning systems operate at approximately 1,000 Hz, producing sound levels between 97 and 112 decibels.
Santa Monica’s municipal code adds another layer of complexity, prescribing exterior noise limits of approximately 50 decibels during the day and 40 decibels at night.
The continuous operation—with vehicles reversing dozens of times hourly, including during late-night hours—continues to challenge community peace.
So, constant car screaming BACKUP BACKUP BACKUP I'M BACKING UP HERE right outside your residential window. Kinda sucks. A whole lot.
Don't worry, we're missing out on a lot of "progress" on this side of the ocean thanks to Trump's dislike of wind farms and RFK Jr's whole anti-vaxxer thing
The setup at SFO is currently quite annoying (Lyft/Uber require you to walk 5 mins to the garage roof, and drivers need to park/wait 5-10 mins away, so there's always a substantial delay). Taxis get the privileged parking spot immediately outside arrivals, but if it's busy you might still need to wait a bit.
I've been wondering for a while why Waymo can't offer a semi-managed solution to SFO to dynamically manage load, have just the right volume of cars inbound, maximize parking utilization, etc. with all of the nice intelligence that an app-based system would enable.
It feels like you should be able to have a buffer of cars waiting right at the curbside, and automatically refill that buffer on short notice depending on observed or predicted demand.
As an Uber rider, I actually love the SFO setup. The walk is short enough, there's actually enough space even during most busy times that there's no crazy honking of drivers trying to get in or out of the pickup zone.
Compare that to the mess that is Uber pickups at JFK, where you have big delays _and_ very poor traffic controls in and out of the pickup zones.
>I've been wondering for a while why Waymo can't offer a semi-managed solution to SFO to dynamically manage load, have just the right volume of cars inbound, maximize parking utilization, etc. with all of the nice intelligence that an app-based system would enable.
Uber could in theory do all those same things too, right?
To some extent, but I think it’s easier to have fully automated buffering if you physically control the cars. Eg you can have backup vehicles parked indefinitely nearby if you want, whereas there will always be some unpredictable churn from human drivers eg unexpectedly clocking off.
No idea if these are first-order effects in practice.
counter point, I love the taxi setup, I wander out, no pre-planning, walk across the street with my headpones on and get in a car, my company pays for it. I suually pay more on uber or lyft, and it's faster and I don't do anything but walk from the plane to the car
I did this in Eastern Europe one time, ended up being made to pay 60€ for a ride that's 10€ on a ridesharing app (even with the "licensed taxi" option...) When there's reasonable price controls it is convenient though
I'm surprised and incredibly impressed at this announcement. It seems trivial, but the general feeling in the industry has been that SF would fight tooth and nail against robotaxis at SFO.
Most likely both agreements had been in negotiations for a while and not something they just pulled together last week in response to SJC, although it's possibly they could have used it as leverage (hey we've talking to SJC ...)
And Waymo doesn't currently operate on highways for passenger service (I think they have authorization to, but they're only testing on highways right now).
They should be able to get to SFO from Millbrae Ave and San Bruno Ave without getting on the highway proper, although it'll likely be a lot slower unless you're getting a ride from nearby. While SJC can serve downtown SJ and Santa Clara without getting on a highway.
What likely happened now is that SFO got a kick up their backside from the Mayor after the press started asking why it was still dragging its feet, while SJC approved Waymo swiftly.
Recent changes in the composition of the San Francisco Board of Supervisors (i.e. Peskin being out of government) may have something to do with it being easier than expected.
Waymo got approval for SJC last week. That probably accelerated approval for SFO, which had been stalling. Nice.
When they get clearance to drop people off at the main terminals, that will be more convenient. Pickup at the terminals is harder. There will be a need for a staging area somewhere in the parking structures.
Few major airports I've been to allow Uber/Lyft anywhere near the pickup area, so many fliers are already accustomed to walking a quarter mile or so to their rideshare. But their inability to use the drop-off area is a new inconvenience, and I can see it limiting the appeal.
Waymo will probably get access to the drop-off area after a while. One step at a time seems to be the Waymo way.
Waymo at airports could work really well with automatic dispatching. They already have an app running in the customer's phone. It should be aware of when someone with a reservation gets off an airplane, and how close they're getting to the pickup point. With good coordination, as the customer heads to the arrival lanes, a Waymo pulls out of short-term parking and heads for the meeting point.
A few more years, and humanoid robots will put the luggage in the trunk.
I'm talking about Uber/Lyft drivers being required by many airports to pick up away from the normal pick-up area, usually down the road a bit or in a parking garage.
It’s wild that $goog is so undervalued (p/e 27) given Alphabet owns Waymo in addition to everything else, and yet Tesla is so overvalued (p/e 243!!!) despite zero Robotaxis in the near (or far) future and lackluster sales.
Goes to show empty promises and fraudulent showmanship sell better than actual working products that people use.
GOOGL is up like 25% over the last few weeks after they resolved the DoJ lawsuit about Search bundling. Clearly there were some investors who thought that was a material risk to the business.
Tesla is clearly a meme stock though, and an example of how the market can say irrational longer than you can stay solvent.
That's why passive index funds and a hands off approach are so often recommended. You cannot really mess up much by buying the whole market and then sitting on it long enough.
Saying it’s “right” based on outcome alone is like saying ~half the people in Vegas and betting on black made a good decision. You can win and still have made a poor decision.
A better approach is to look at the full range of your bets and try and decide if the betting strategy was good. But that gets difficult when you consider outcomes are linked through wider economic trends.
If you buy Alphabet stock you're betting on the whole company doing well.
Google makes around $300B a year. Uber's entire business makes around $50B and that took a decade. Waymo would have to become a major business to move Alphabet's stock price in the near term.
Considering Waymo is very likely losing money, experiment very slowly with scaling up, and still raising billions in private capital outside Google... idk. Doesn't seem as simple as buy $goog in 2025.
I think Waymo has huge potential for being much larger than Uber - people are willing to pay more compared to ordinary uber drive just to avoid dealing with taxi drivers and tech will only get cheaper.
More than that, I think the ride-hailing business is just the fist volley in the self driving vehicle space. It’s a short jump from there to self driving trucks, self driving package delivery, self driving private vehicles, and on and on.
Can any of those companies catch up on self-driving faster than Waymo can pivot to their niche? Cruise seemed to be a distant second, but did themselves in with an attempted cover-up.
> There are already self-driving trucks on the roads.
2 trucks?! I suppose that's the minimum number required to make your pluralization correct.
I will stand on my earlier statement regarding this particular outfit: they'll need to catch up because Waymo started class 8 variants in 2021 https://waymo.com/blog/search/?t=Waymo%20Via
I see Australia in the article and pardon my rampant scepticism, simply don't believe it.
Lo and behold:
>A six-month trial of driverless trucks on public Victorian roads has been put on hold just hours before it was meant to begin after the transport union labelled it “shambolic” and “sneaky”
> "the futures of our truck drivers are jeopardised due to this poorly executed plan."
> “It’s unacceptable that these trials are being pushed by corporations that continue to disadvantage our hard-working mums and dads that work day in, day out to carry Victorians.”
Now this sounds far more like the Australia I know.
Looks like the entire trial was scrapped due to union pressure and never resumed. Same reason we can't even have Driver-Only Operation on NSW trains, despite specifically purchasing DOO trains that operate safely worldwide.
And plenty have failed. Perhaps a smaller problem space but still really, really hard. Some self driving freight company failures: Starsky, TuSimple, Embark, Ghost, among others.
One promising self driving truck startup, Aurora, was forced to put a safety driver back in the driver's seat after testing in May.
"Forced" by the truck maker, who was forced by their insurance company. All these companies will face that hurtle. I suggested to my girlfriend, who is a corporate defense attorney, that she get involved in this area of legal practice. It's a legal minefield.
Cruise was nixed by GM execs, whom I believe were looking for whatever excuse they could find to shut the operation down. They simply couldn't afford to stay in the game for the long haul. Cruise was under pressure to appear more capable than they were, and they took risks.
Waymo is distinguished in that it doesn't need to pander to nervous investors to keep getting money. The company is Sergei and Larry's baby. Google's founders will ensure that Waymo is patronized until it can stand on it's own.
> ...I believe were looking for whatever excuse they could find to shut the operation down
Cruise's self driving license was suspended because humans displayed poor judgement by omitting from the official report details of their stopped car dragging a knocked-down accident victim under the car for dozens of feet. They took "risks" alright, and their harebrained cover-up was discovered by chance by the oversight body.
I believe any driver who covers up the details of injuries in an accident permanently lose their license, because they'll definitely do it again. What good is a self-driving subsidiary that can't operate on public roads?
Buy a Comma.ai and install it in a supported vehicle, and just try it out. It doesn't talk to GPS, but it handles left right gas brake on the freeway well enough, and that's with two fairly shit optical cameras and a radar system. Granted, geohot helped start the company, and he's no slouch, but if their system is that good, a couple things are true. A) Lidar isn't necessary b) Extensive mapping that Waymo does also isn't necessary c) that last 10% gonna take 500% of the time to get to L3/4/5 autonomous, and that last 1% is maybe never. The other day I was in a Waymo, and there was a semi totally blocking the street, backing into a loading dock. The Waymo correctly identified that there was an object in the way, and stopped and did not plow into it. At first it crept up to the semi, blocking it from making progress as well. It might have started backing up, I've seen them do that, but I was already on the customer support line as soon as I saw the semi blocking the road.
Comma.ai is probably the purchase I'm most happy with this year (to be fair though, I buy a lot of crap off Temu). Drives are now just "get on the freeway, and just chill." Pay enough attention because it's not collected to GPS and just in case something goes wrong. So to be clear, Comma.ai is not autonomous driving, it's classified as an ADAS, advanced driving assistance program. It just makes driving suck that much less, especially in stop and go traffic, for $1,000, and compatible with recent vehicles that have built-in lane guidance features. Waymo's got to be light years ahead of them, given how much money they've spent, so it's my belief that Waymo's taking it very slow and cautious, and that their technology is much more advanced than we've been told.
There are several “last meters” delivery robots developed.
Short range drones are being used in Australia.
And I heard of at least one company working with apartment architects to standardize a “port” on the building exterior to which a truck/robot would connect to “inject” packages to the inside.
> "Short range [delivery] drones are being used in Australia."
Last I read (late 2023 IIRC) these were being cancelled in various areas, if not everywhere? People in neighborhoods were getting annoyed by the noise of drones buzzing overhead.
This was just an acquaintance some years ago in SF, but I recall it was fancier with conveyor belts and a protocol for the robot to communicate the size and weights of the packages being delivered.
But to be serious, there may be a way of doing it, it just seems very far off unless you're talking about Amazon hub or something like that, where it would be more feasible (but still difficult to achieve).
Think of Waymo Driver as the equivalent of Android for vehicles. It's an operating system and a suite of cloud services for both autonomy and ride hailing.
What about all the expensive hardware, gpus, lidars? That’s like having iOS on your phone and if you want android you need to buy extra things that are worth same price as your phone.
And costs should be lower in the long run if you don't have to share the ride fee with a driver (not case yet because seems like they still have alot of staff to manage the cars)
Statistically Waymos are more expensive than Uber rides, but practically as an individual they are often cheaper than Uber, its very easy for the stated price to be lower
You might, but most people wouldn't, and more to the point, overwhelming more people will choose to drive their own car (or take transit) vs either Uber or Waymo.
If Waymo can drop its price by 50%, it could steal a lot of demand from normal cars and transit, but that doesn't seem like it's even on the conversation right now.
I would need to see Waymo be able to handle something like Southeast Michigan before I could even get comfortable with trusting it to get me ubered t/o from home for maintaining the vehicle I need to commute when I can take a remote day or two...
And then also delivering that for a good cost.
I put it that way because, I do tip Uber drivers well (unless they cray cray) and they would need to properly 'undercut' uber with whatever model they serve up in more complex areas.
Why is southeast Michigan difficult to drive in? I don't know anything about the area but I would guess if GPS navigation works and it's less dense than SF/LA, most of the major issues are solved?
Anyone who's taken enough Ubers and/or has had bad enough luck to have gotten a terrible Uber driver. Pretty much everyone I know, along with myself have had multiple awful Uber driver experiences.
Did uber/lyft get radically better in the last 12 months?
I had one rapidly cycle their prius between 50 and 70 on the freeway because regenerative brakes save gas (I felt carsick for hours after arriving at my destination), and another actually get an angry mob to tap on the windows and berate their driving. (The mob was justified.)
Since then, I’ve given up on using them whenever possible.
> rapidly cycle their prius between 50 and 70 on the freeway because regenerative brakes save gas (I felt carsick for hours after arriving at my destination)
Weird take to me, unless you were on a lot of hills; at least in my Maverick [0] 55-65 is 'ideal' MPG range for long trips, going between speeds tends to trip things up and actually -avoid- the weird 'battery has enough juice where we just kinda lug the engine' mode.
Doing regenerative 'braking' compared to using physical brakes, absolutely can give energy for momentum/acceleration and save on the physical brakes wear and tear, OTOH any normal cyclist would say it's better to 'maintain' a given output power vs allowing deceleration and then going back up to speed.
As for why, well I'm not a physics person, but in general it's that you are having to overcome the rotational mass/etc of the wheels (i.e. tires, axles, etc), and no regenerative braking within the current laws of physics will make slowing down and speeding back up more efficient, at least on a flat road.
[0] - OK It ain't quite a prius but it works fairly close aside from overall drag...
That TC article doesn't substantiate its overly broad claim. "People" aren't paying more, in general, across its US markets; it only shows that a subset of its customers in what is already the top-5 most expensive cities (SF) in the world are prepared, and at that, only 10-27% are prepared to pay significantly more ($5-10). Still fewer than the 40% who would pay “the same or less.”
Quoting: "Perhaps even more striking is how people answered a question about whether they would be willing to pay more for a Waymo. Nearly 40% said they’d pay “the same or less.” But 16.3% said they’d pay less than $5 more per ride. Another 10.1% said they’d pay up to $5 more per ride. And 16.3% said they’d pay up to $10 more per ride."
There are going to be lots of causal factors: number of rider(s), time of day, safety, gender, wait time, price estimate, predictable arrival. Let's see an apples-to-apples comparison/regression breaking out each.
I think waymo actually has a better km/accident ratio than the average driver. Plus if you haven't done it before, it'll be a cool experience to ride in a car with no driver!
But in the long term I think the point of waymo is that it'll be cheaper: no need to pay the driver if there isn't one!
The words women and woman appear exactly once each on this thread. If there's one thing tech product management needs, it is to ask a woman. This is the most obvious blind spot in tech.
Maybe with the HN readership, but in general the public don’t want to drive in driverless vehicles and don’t want them on the streets. It’s going to be a long uncertain road for them to be accepted.
I don’t think Waymo is very likely a losing money experiment. I give them a 50% chance to be successful within the next 10 years. Successful being that self-driving cars are able to operate in 50% of the world/terrain types/region types, probably within another 10 years to scale up.
To all: also think of the productivity boosts. Working in the car or just napping in the car.
In the Netherlands this is already sometimes possible if your work is close to a train station while your house is too and you don’t need to switch trains. It’s a boon to be honest.
My favorite is the train from Amsterdam to Berlin.
Of course, if you carpool then you can do this too. One time I rode in a car as a passenger from Berlin to Prague while working the whole time. When we were there, we went to a DnB festival and we got back on the weekend.
They have already spent an enormous amount of money. It’s hard to see how they could make it back quickly, if ever. I’d like to be wrong, but I expect they will continue to be a money losing experiment for a long time yet.
How much money they've spent in the past is irrelevant. That money all came from investors, in exchange for a stake in the company. It never needs to be "paid back". Besides which, those investors have earned all those funds back already, and then some (on paper).
All that matters at this point is how much money they'll lose/earn in the future. There are no shortage of investors willing to put money into this effort, and they're growing exponentially, so there won't be any pressure for them to turn overall profitable for several more years.
Boeing may never make back the development costs of the 787. That was an absolutely epic disaster of a project. But that doesn't mean Boeing shouldn't build and sell every 787 they can profitably sell.
If Waymo is at breakeven including capex, opex, and overhead, operations logistics becomes the limiting factor. While Alphabet is capable of investing more money into Waymo, I think they've reached the tipping point. If you see Waymo expansion accelerate, bet on that tipping point having been reached.
How much money do they make off the average person in the value of ads shown per year?
Now compare to how much money the average person spends on driving per year.
If Waymo winds up running half the market in autonomous transportation over the next several decades, it'll make search look like peanuts in comparison.
You need to consider profit margins. The cost of showing somebody an ad is very near $0, which is what makes digital products so profitable. But when you do things in the real world, especially in highly competitive markets where the customer is extremely price sensitive, your profit per mile is going to approach $0. For instance WalMart's profit per item sold is less than 3%, and for driving this will likely be substantially lower (given the combination of customer price sensitivity + competition). The way you make up for this is in massive volume, but Waymo for now remains a heavily ringfenced operation and so it's not entirely clear how they reach scale. Google also has a very poor record of long-term performance in competitive markets.
The winner in self driving will likely be enabled by extreme vertical integration - you want to be building your own cars, cleaning your own cars, repairing your own cars, and so on.
The average American sees something like $500 of ads that go through Google per year. There's a profit margin of around 50% since Google has to pay publishers and pay for running search. So that's $250/person in profit per year.
The average American spends something like $12,500 in car+taxi/rideshare per year. Suppose with Waymo that goes down to $7,000 and it's 20% profit. That's $1,400/person in profit per year.
Obviously it gets much more complicated -- the profit margin depends on whether there are serious competitors to Waymo and how much Waymo's head start matters. Waymo will bring costs down further with shared vans and buses on demand. Profitability will rise with video ads in vehicles that you pay not to see. And so forth.
But autonomous rideshare is going to be larger than search any way you look at it. Profits won't be as high as search, but the barriers to entry are so high that profits will be high for a long time.
Those data you referenced are per household, not per person, and the majority of that is loan+insurance. The actual cost in terms of maintenance, fuel, etc is quite low, and that's the price that eventually will be the goal line for autotaxi companies. 20% [net] profit margins do not generally exist in competitive real world industries, outside of perhaps something like real estate. A net profit margin of 5% would be huge, and I think it will likely be much closer to 1%, or even less, simply because in the end it's going to be a commodity where all that matters is price.
I also think you're overestimating the impact of things like ads, buses, etc. The second Waymos become less pleasant than any remotely comparably priced option, they will lose customers.
No it's per adult not per household. The average household has 2.2 cars, so the figure per household is much higher. And it doesn't matter what proportion is loan vs insurance vs maintenance vs fuel, because Waymo replaces literally all of it.
And yes I assume Waymo will have high profit margins for an extended period of time because they have such a massive head start, and for a long time will be competing primarily against rideshare with human drivers, so won't be pushed below that. Their marginal costs will be much cheaper than that, not having to pay drivers. Hence 20% is not unreasonable.
Then, even in the long term, the economies of scale they develop and network effects will continue to give them a significant advantage. Not 20% margins, but way more than 1%. Especially as they start to vertically integrate the hardware at some point.
Here is where you would generally cite sources. [1] Those are the data from the BLS. Total transport spending per household is $13,174. The term they use is consumer unit, which you may have conflated with consumer/person, but it's practically the same as household. There are 134m consumer units, and 131m households.
Waymo is currently charging substantially more than Lyft/Uber and is not profitable. Human drivers can taxi in anything with 4 wheels and a hood, and its 100% their responsibility to take care of their vehicle, fuel it, clean it, and so on. Each Waymo currently costs ~$200,000 and is going to have a proportionally higher maintenance costs, and all of those costs must be covered by Google. So their costs are far higher than you're ballparking.
As for competition - Tesla has already launched a live robotaxi trial in Austin, so it's already here.
I was just going off the top Google result based on AAA data. Took a closer look and it turns out it's the average for new cars [1], so the discrepancy must be that your statistic takes into account the secondhand market. Thanks for the correction.
In any case, the overall point is the same -- it's a vastly larger market than Search. And what Waymo currently charges, and the current cost of their cars, is irrelevant. Waymo's business model isn't based on the economics this year or next year. It's based on the economics ten and twenty years from now, when costs have fallen dramatically as they switch to cheaper models and gain massive economies of scale.
As for Tesla, it's hard to take seriously given all the promises it's made and completely failed to deliver on. Their trial currently has a safety human in a front seat and is limited to a tiny group of testers. It's so many years behind Waymo already, and it's unclear if the technological approach it's taking will ever be able to catch up or meet minimal safety requirements.
Can’t imagine Tesla will be able to remove the passenger seat safety monitor any time in the next 5 years. Refusal to install lidar means Tesla’s AI has to be 100% perfect, which won’t happen for a long time, if ever.
When these are ubiquitous enough, the vast majority of people who currently own cars won't need to. It'll be so much cheaper and easier to use rideshare.
I can't really imagine the circumstances where I wouldn't want to still own my own vehicle even if it had an autonomous mode. I drive it places where I don't have cellular service. I keep lots of stuff in the vehicle. It's customized with accessories like roofracks. I can hop in my vehicle from my house immediately whenever I want to.
If I lived in a city and garaging a car were inconvenient/expensive? Maybe. But that's not me or a lot of other people.
But if it's half the price over the course of a year? And you can summon it in advance cheaply? And it basically never takes more than 5 min to arrive anyways, since they're everywhere?
You might decide it's worth it to keep the stuff you really need in a messenger bag or backpack or something, the way people in NYC do. And maybe roof racks don't matter if you can just summon a second autonomous van behind you to hold whatever you were going to put on your roof.
Obviously if you're a contractor or something you'll need your own vehicle. But the point is that for most people, sure they can't keep stuff in their trunk all the time, but that's a happy tradeoff if the total cost of driving is 50% less.
Of course there will be exceptions or holdouts, but it will come for gig drivers, then for second cars, and go from there. There will be versions with roof racks, with extra luggage space, with child seats.
Statistically, and from a global perspective, the apartment-dwelling car owner (most likely with a lower income than yours) is a heck of a lot more common than living in American-style suburbia or a small town.
Uber took 14 years to make it to profitability. Money's frequently characterized as impatient, unable to look past the next quarter, but when it wants to be, it can wait.
Waymo's older than Uber, but they hold many key patents by this point. Now that they've started running a taxi service, it seems straightforwards to scale up, assuming that is the business they want to be in. Then it's just a matter of charging more than it costs to run the service, and wait.
What makes investors patient when no profits for years, is when they see growth, entrenched commanding lead and network effects, large user base etc. As long as investors can imagine a good likelihood of eventual profitability, then growth in the present is a fantastic substitute for profits.
Growth tells you the eventual profits will be bigger. Leadership and moat gives certainty that the company will actually get the profits for the market they grew.
Imagine if you could buy your own "Waymo-equipped car". No need for driving lessons. No aggravation. No road rage.
How many people would pay for such a luxury car? With the US population aging and public transit non-existent in most places, Waymo probably has a market for cars.
There’s clearly a demand for self-driving privately owned vehicles as well, but think of it this way - why own a self-driving Chevy when you could hire a self-driving Cadillac when you need to go somewhere?
Uber making 50B, probably means Uber is paying drivers around 200B or higher. So that is Waymo’s potential revenue in the long term as it releases in most ride share markets. I think it’s under 1B revenue now, which just shows how much growth ahead is possible. Even if we think Uber will be at least 50% market share in the coming decade, at least 100X growth is left for Waymo. This also completely ignores Waymo creating latent demand, which is wholly possible. I would for example trust a Waymo to drop my kids everyday over an Uber.
The cost of the computers, LIDAR, special maintenance, vandalism, staffing humans for remote issue handling etc will probably costs the same as a year's income for an Uber driver. But after that it's mostly profit and they can run cars longer.
The most important thing for Waymo is scaling up production of LIDAR and maintaining them efficiently. They will have a massive fleet running very sophisticated radar+computers. That's a huge logistical investment when it's a million cars. Those sensors will break or be damaged.
They've been partnering with Uber to maintain the fleet in some cities haven't they since they already have regional infrastructure? I don't think they want to be in the fleet management business.
AFAIK Uber is doing app integrations + some local operational fleet management. Waymo is supplying the cars, radars, computers, remote service, the brand, etc. Waymo has to scale that production and maintenance up country wide and then globally.
Uber's CEO compared it to Marriot, people come in to run the hotels in the local region, but they actually don't own the hotels. It's like hired managers who take a cut.
It also makes sense to have people with local experience run them in each local region. But those businesses still involve margins and expenses that have to make sense.
Don’t forget that Waymo will always be a much lower margin business than search! Setting aside the decades of R&D expense, those cars require purchasing, maintenance, warehousing, etc.
All that may be true. Human drivers are not the point of comparison. The search business is. Waymo will still always be a lower margin business than search for the reasons I enumerated.
Waymo may end up being great business. But it is unlikely to exceed what search is/was. For that reason, press X to doubt GP's claim that Alphabet is undervalued. "IT'S PRICED IN" [1]
It's unlikely to exceed what search was, but transit is a much more reliable bet for continued revenue. I don't think anyone is betting on ad revenue being reliable at Google long term anymore.
But the market is so, so much bigger. And the margins will likely stay high for a long time while there are few competitors, and their main competition is human drivers.
Not having to pay drivers is an enormous source of profit.
As big as search!? Doubtful. The entire globe is unlikely to be the addressable market. China will never let Waymo in. India will undoubtedly field multiple worthy competitors. Europe is hostile to technological progress and even more-so to American tech cos. In most parts of the world, Waymo is unlikely to be able to deliver a positive gross margin business given the per-capita-income of most places.
It could be a big business. In fact, I hope it is. Lives will be saved. But there is still a lot to be worked out, and the margins will never be as sweet as those of search.
One of the main reasons to vertically integrate is to expand margins by squeezing cost out of the value chain. My point still stands: Waymo will never have margins as good as search.
He has done many impressive things, but one consistent thing about the man is that he always over promises and regularly under delivers. The examples are too numerous to count (smashing the CT's "armour" glass, humans to Mars in 2024, Thai cave submarine, naming your driver assistance technology Full Self Driving, etc, etc)
Perhaps that's simply the price of achievement, but Showman is apt
But yeah I didn't realize Waymo's coverage is more than Austin and SF where Tesla rules already. So maybe end of year they'll overtake. Which is crazy Waymo is sitting on this. Even at 10x more expensive cars you'd think they would just put their cars everywhere, but scalability bottleneck seems to be software or lack of remote ops.
Don’t forget Zip2, PayPal, Neuralink, OpenAI, and The Boring Company.
There are large swaths of people that accept headlines as fact and/or cannot or will not grapple with nuance and complexity (“I think Elon’s a jerk and he is a formidable engineer.”) Perhaps it’s a sign of these polarized times, or, as I believe, people have always been like this. We just have more time and resources to dedicate to outrage and flamewarring than we did in the past.
Yes, there has been nice geniuses (ie. people with extreme talent), Mozart was for example a good person. Da Vinci (if a little sycophantic when young) was not unhinged at all nor abusive and was appreciated.
But since romantism we have built this image of the genius as necessarily abusive.
I’m sure abusive genius are very visible (by definition?) and that abusive people tend to monopolize more ressources too. (Like these tenured professors that use their students to advance their own career)
I think you guys shouldn't be comparing “geniuses” because i don't think thats the forcing function here (ie IQ and ability).
The forcing function is having so much responsibility and stress from running so many companies. You have no extra bandwidth for anything. All your time is spent.
So maybe look at comparable people with insane schedules/workloads/very high pressure situations.
Fair. With Elon it feels like there's an obsessiveness that drives him to take on so much responsibility. And as you say, that can affect what he says publicly.
True -- also I wouldn't say Elon is a genius. I feel thats a term for people who solve deep intractable physics/math problems. Elon's admirable attributes are that he is an insane capital allocator, has a very acute engineering mind (rare for leadership), curious mind, sees the future paths, dedicated focus and is an unabashed salesman of his products and philosophy (maybe this one isn't as admirable but its critical to his success).
He was ousted from Paypal before anything major happened, he was basically just a shareholder.
The Boring Company is an obvious bust. So is the Hyperloop. Neuralink is another likely bust. Tesla solar is going nowhere. The Cybertruck is a millstone around Tesla's neck. Etc, etc.
Usually Elon's technical flaws aren't on display, or at least he covers them well. For example while it's true FSD hasn't worked out, but I don't know you could say at the time "most competent AI devs knew it wouldn't work out". However, when Elon attempted to move PayPal from Linux to Windows, most competent software engineers would have advised against it. Paypal isn't an example of Elon's genius in action - it's the opposite.
When Tesla introduced HW2 it was clear to people in the self-driving industry that it wouldn't work out. Elon was insistent on repeating mistakes that other companies had already learned from. Of course the other companies never considered some people's willingness to pay good money just to pretend that their cars can drive themselves.
Says who? I've tried it and the capabilities are amazing. If you told me 10 years ago that I would be able to buy this in 2025 I wouldn't have believed you.
Says me, who owns a Y that has the FSD package. Random braking on a highway, indicator lights coming on for no apparent reason, windscreen wipers the start on a dry day, attempts move through a red light. None of those things are common, none are serious if your hands are on the wheel and you are giving it your full attention. It's a serviceable attempt at FSD Level 3, and auto park works well.
But when I bought it, Elon was promoting hiring out your car as a FSD level 5 taxi when you weren't using it. If I regularly took my hands off the steering wheel and went for a snooze (if that was possible, which it isn't because they would be sued within an inch of their life), I'd be dead by now.
I think the real purpose of the Boring Company and Hyperloop were preventing/slowing expansion of public transit, and that by that measure they were successful.
I don't think it was a carefully calculated conspiracy (such as 1)
I think it was an engineer with found wealth starting to do stuff with it.
but nowadays I think he has evolved into something different, maybe some of it from the wild public feedback loop, some of it because some of the things he cares about are going wildly wrong.
There is nothing in the article, the twitter thread it quotes, or the text from Musk's biography quoted in the respective tweet, that indicates that the Koch brothers assisted Elon Musk in any way in trying to sabotage California's high speed rail. They're simply mentioned as other people that oppose transitioning away from automobiles.
Furthermore, Elon Musk doesn't say that the Hyperloop "was a conspiracy designed to sabotage high speed rail." He is quoted in his biography as saying that he hates high speed rail, doesn't want them to build it, and thinks it's a waste of money. He also says that he had no intention of leading the effort to build Hyperloop himself, where he's directly quoted as saying, "Down the road, I might fund or advise on a Hyperloop project, but right now I can't take my eye off the ball at either SpaceX or Tesla." The biographer speculates that this means it was a cynical ploy to get HSR cancelled, and I don't think it's unreasonable to infer this, but one could just as easily infer that Elon really did want the California legislature to build something akin to a Hyperloop instead.
There's no debating that Elon hates public transit, he'll tell you himself[1]. You don't have to spread misinformation to make that point
I am "just a shareholder" in Paypal. Elon Musk had a > 10% stake inherited from his ownership of one of the companies that was the precursor to Paypal itself. It's not remotely the same thing. And listing failures is not meaningful at all. Failure is the default outcome in business.
Either go ague with Wikipedia, or put some argument in the comment when making claims you expect people to verify themselves. People are just going to look it up on Wiki.
> SpaceX was founded by Elon Musk in 2002 with a vision of decreasing the costs of space launches, paving the way to a self-sustaining colony on Mars.
but if they're google's products how would they cannabalize ads biz. would revenue not just shift? or do you believe ai search will be overly adopted but not as profitable?
Google was late to search, late to smartphones, late to internet email. I'm having a hard time thinking of any of their large markets where they were a first mover, maybe YouTube-ish, widespread user uploaded internet video wasn't meaningfully available before the rise of YouTube.
On topic, Waymo is clearly a first mover in self-driving, having the first legal commercial services.
But, being the first mover is usually more of a disadvantage than an advantage, IMHO.
I'm struggling to think of a single product where the first mover won. At best they are able to hold some market share like Dropbox or Slack, but eventually big tech moves in and crushes them by just offering the same thing but cheaper and more integrated.
I believe TSLA also represents their humanoid robot segment with some questionable addressable market definitions done by investment analysts. I believe it’s overvalued but they are a forcing function for the other tech companies to push ahead
Waymo is a small portion of Alphabets business, while cars are a massive portion of Tesla's. If waymo was seperated out from Alphabet maybe it's p/E would be that high.
TIL. I stand corrected. Though worth pointing out (as the article does) that on September 1st, new legislation in Texas was passed adding some restrictions to autonomous vehicles. So seems reasonably likely this is more regulatory than necessary.
Because some people read beyond headlines and realize that Tesla will most likely dominate with Robotaxi. Their traditional consumer vehicle revenue could pale in comparison. And Optimus could be another order of magnitude larger.
That’s the optimistic bull case. It’s not impossible.
Tesla will be able to scale Robotaxi much quicker than Waymo can scale.
Why? In principle the basic Waymo technology could be adapted to work on any modern vehicle. They aren't dependent on Jaguar manufacturing capacity to scale up.
It's capital intensive to make all of those devices. Tesla's strategy is to rent back devices they sell to consumers. This lowers the necessary capital costs and will enable quick scaling. It's a similar ploy to how Amazon quickly grew its delivery capabilities.
Tesla still has no autonomous vehicle that customers can actually buy, let alone rent back for taxi service. So any "strategy" remains entirely hypothetical.
>Alphabet has $95B of cash and short-term investments
Not only that, but also they could probably raise 10 times that much by creating new shares and selling them (if they had a plausible story to tell investors as to why the money would be well spent).
I so wish Tesla had gone Waymo’s route and focused on delivering really safe LIDAR-based level 4 with cheaper hardware. I think they’d be well-positioned to take the market by storm.
But instead they made an ideological stand on cameras only, and they’re helmed by an unhinged drug addict who lies constantly, to the point many who once would have loved to buy an actual self-driving Tesla now won’t touch them with a ten-foot pole, even if they do someday deliver an L4 experience.
I really, really hope Waymo licenses their tech. I think that would stomp Tesla into the ground once and for all.
it's bizarre how even the smart people here (smarter than me on average, no doubt), think they know better than Karpathy and Musk who have spent a decade deep in this problem.
SpaceX has landed orbital boosters 500 times, and STILL no one else has done it.
Teslas drive millions of miles autonomously (yes supervised, but still) every day.
You can't even type a response without containig your political/social bias on anything related to Musk.
Tesla is still at level 2 autonomy, whereas Waymo is at level 4.
It doesn't appear Tesla can achieve level 3 autonomy given Musk's ideological opposition to lidar. Without lidar, the AI has to be 100% accurate, and it's not and won't be for the foreseeable future.
As they say in Maine, "You can't get there from here."
The level 2 vs 4 thing is a technicality and fairly misleading, Teslas can already operate in a much wider range of roads and conditions. Waymo is still on guardrails. Mercedes touted "level 4" but you had to be following another car, going slow, etc etc. General autonomy is what matters.
I trust Karpathy when it comes to lidar vs vision. Do you shoot lasers out of your head to drive?
LIDAR sensors are getting cheaper faster than camera-based autonomous driving software is improving. I predict that in a few years, regular luxury cars that are still mostly human driven will come with LIDAR for collision avoidance and improved driver situational awareness. Just like they already use RADAR for the same purpose today.
Not a problem. The costs for sensors always falls rapidly as production volumes scale up. The first GPS receivers were large dedicated devices that cost thousands of dollars. Now they're just a chunk of IP in the SOC for every little consumer device and cost pennies.
My Model Y in Vancouver drives me to and from work daily. I cannot get a Waymo here -- and I certainly cannot purchase one privately. Which is more effective where I live?
I don’t doubt that Waymo car is more advanced than FSD, but that comparison isn’t as impressive as it sounds. The numbers of FSD equipped Teslas dwarfs that of Waymos, and they are available everywhere, not just selective cities. You have to take that into account.
Teslas is also much cheaper, and easier to scale. Tesla has better growth potential even if their tech is less impressive.
It's not that their FSD tech isn't less impressive, it's that it's not FSD tech.
Even worse (for Tesla) is that if they do try an make their non-FSD tech do FSD, and it decks little jimmy because the flashlight in his hand looked like a far off street light, Tesla is liable to face a knee-jerk federal law mandating lidar. And just like that the dream is dead.
This forces Tesla to be extremely paranoid, as it's one visual mistake away from being told to use lidar.
Why is a 34x improvement in the rate of interventions not as impressive as it sounds?
I’m not even sure that Waymo number is still correct. They’re doing hundreds of thousands of paid rides per week, with no one in the front seat, so not sure what an “intervention” even means at that point. Maybe where the passenger needed help and called support? That’s 1000x better than needing to grab the wheel because your Tesla was about to drive into oncoming traffic or run over a kid in a wheelchair.
We’ve also not seen how capable Tesla is at evasive maneuvers. We have plenty of videos (hundreds now) of Waymo making instant swerves to avoid children running onto the road, cars running red lights, a person falling from a Scooty etc. These are not maneuvers you would expect from a human, which shows how Waymo has pretty successfully crossed the human bar in safety. If Tesla does not demonstrate this, on top of driving normally, I don’t think they have a product. The barrier to give control to a computer is super human not human like driving.
Also philosophically I don’t see how a big neural network will create such evasive maneuvers, unless you try to create such scenarios in a simulator and collect evasive data. Seems prohibitively expensive to do so in the real world.
Overvalued by traditional (PE) means. I've ridden in Waymo (50+) and Austin Robotaxis (12). Tesla has Waymo beat in terms of human-like feel, interior features (sync to your own Spotify, Youtube, etc). When Tesla removes the passenger seat monitor, scaling will happen much faster than Waymo... Tesla just received the initial license for driverless Robotaxi in Nevada. Tesla also produces more Robotaxi-capable Model Ys in ~6 hours as Waymo has cars in service (in total).
Tesla's self-driving technology is a joke compared to Waymo's and the Tesla brand is extremely toxic now. I see from your other comments that you're big on Tesla (own several and have a son who works there) but as an unbiased observer I cannot fathom them winning this market.
I have 2 AI4 Teslas with FSD, and I don't find V13.2.9 lacking at all in the Vancouver area. V14 will be a 10x increase in parameters, too. Why do you feel it's a "joke"?
It's a "joke" (I wouldn't call it that, but it's a vastly different product) because you have to pay attention to the road at all times.
You don't live in a Waymo city, so I understand. A lot of people who don't live in a Waymo city don't really get it.
Waymo is a completely different product than FSD. It's a robot that comes and drives you from point A to point B. You can do whatever you want while it's driving, such as take a nap or work on your laptop.
Tesla was SAE level 2 in 2013, and they are still SAE level 2. Waymo's Robotaxis are SAE level 4, and they can drive on public roads empty with no human supervision, both technically and legally.
I have friends on the Autopilot team (and a son). Their goal is by end of year. I've been on HN for 15+ years, and seemingly the only downvotes I get are when I post my thoughts and opinions on Tesla.
1:1 is going to be ruinously expensive. You need three shifts of remote operators. Even in the Philippines or Vietnam, if you can make the latency work, that's prohibitive.
> How do Elon Musk's predictions relate to Tesla achieving a robotaxi service or not?
>
> Ignore his predictions and just... look at whether or not the Tesla FSD team is making progress.
I'm seriously baffled by this comment. How can Elons comments not be relevant? How are you proposing we assess the progress of the FSD team? And why should the assessment be different to the last 5 years where FSD was supposedly ready (according to someone with intimate insight into the work of the FSD team) by the end of the year?
> How are you proposing we assess the progress of the FSD team?
...any metric you want? Miles driven under FSD. Miles driven without intervention. Miles driven without accident. Anecdata from friends of yours who own a Tesla. Whether or not a partially supervised pilot program has been launched in some cities.
If Elon Musk said in 1999 "I think we will achieve self-driving next year", that also has no bearing on whether or not self-driving is achieved in 2025 (in either the positive or negative direction). It only means that Elon Musk's "predictions" can't be trusted as an accurate harbinger of success. Which is precisely why you look beyond his words and at the reality on the ground, which strongly indicates Tesla has made a huge amount of progress in the last 10 years, and could be very close to having unsupervised robotaxi service in various jurisdictions.
If we use kilometers driven with drive assist as a metric then nearly car manufacturers will have robotaxis by the end of the year.
If we talk about anecdotal evidence then I know people who are deeply familiar with the topic (working of self driving technology at other manufacturers) and they say fully self driving is still many years away for all manufacturers. Moreover the general industry sentiment is that Tesla is behind now and that more sensors then just cameras are needed.
But instead I should believe the Tesla fan boys who just like Musk have been raving about the amazing progress and telling me that FSD is just around the corner for years.
Never heard of "universal A-to-B navigation", that sounds like google maps.
Is it fully self-driving, like Waymo? If not, then I'd lump it in with anything else that isn't fully self-driving. Either I can safely and legally nap while commuting or I cannot. Something that requires me to actively supervise the car and intervene as necessary is not self-driving, it is drive assist.
> "What competitors say" is quite possibly the worst anecdata you could find as a broad rule, no?
The post you're responding to is not simply repeating what competitors say, it is speaking of using data to avoid trusting what anybody says. Thus, this isn't a fair comparison. It should also be noted that you yourself suggested that the poster use anecdata.
That said, what tesla says about themselves is even worse than what tesla competitors say, if only because tesla is infamously untrustworthy, and their competitors are not.
But again: don't listen to what tesla et al say they will someday do, compare the data for tesla's drive assist vs tesla competitors' drive assist.
You're not "posting your opinions on Tesla", you're literally shoveling them into everybody's throats. You'd be "posting your opinions" when it was one, two comments, and not plenty, like under this news. You're a Tesla freak or fanboy, not an objective commenter.
I think the downvotes might be due to one or more of the following:
- You're uncritically parroting the notoriously untrustworthy talking points of a notoriously untrustworthy company, and HN posters expect more critical thought in comments.
- You're redirecting to some rumored "goal" rather than a realistic prediction, which was the topic, and HN posters liked the topic.
- HN posters may think that your vested interest in tesla behooves you to think more critically than the average person on matters involving tesla, rather than less, to overcome any implicit bias you might have.
- I have a goal of end-of-month, so that means I'll have it even sooner than tesla, right? This is how many view the claim by tesla, except I, a random person, literally have less of a reputation for dissembling and failure to deliver than tesla does.
I drove for Uber/Lyft back in 2020 and let me tell you, SFO is a nightmare. I missed a turn once and had a passenger trying to make a flight furious at me. I quickly figured out there were a group of drivers who specialized in SFO and amatuers like me should avoid the place. When Waymo announced San Jose I thought ok, that makes sense because SJC is easy, but SFO? Wow, I'm impressed. I hope it goes to plan.
Nothing more rewarding than a company working hard and seeing real-world, first of its kind results in action. Makes me feel giddy about a company again like peak tech back in the 2010 era.
Congrats to the Waymo team, I’m sure this was a huge milestone internally.
Oh, this makes a bit of sense. The Avis/Budget fleet team will be part of managing the vehicles, so they can be quickly cleaned and fueled up when they slide into the airport, too.
Same. I go to the rectal car center at least 4 times each year. I just was there on Saturday and had no idea either. Still don't know what it is other than Waymo pickup.
I did always find the term kiss and fly confusing and weirdly intimate, as if everyone is getting a ride to the airport from a spouse or parent. Definitely a throwback to another era.
ChatGPT tells me it originated at Paris CDG in the 1990s where it was "depose minute" but then also cites LA Times from the 1950s as the first source of the term[1]. I've seen it in Germany as "Kiss und Ride".
So it's not really a regionalism, but I also don't think it's super common.
I'd hope so. As an aside, I wish Waymo was more transparent on the app that their cars are not allowed to take passengers on the freeway. I was unaware of this restriction when I booked a ride from SF to Burlingame last month and I was stuck in a Waymo for an hour going down residential streets!
I wonder how that'd feel. I took a Waymo in SF last fall and I was pretty impressed. But it was also slow city speeds. I wonder if it feels different going at freeway speeds with "no one" at the wheel.
While the margin of error is much lower on a freeway due to the speeds, other drivers are generally a lot more predictable (also in part due to the speeds).
Sure - a good freeway is actually a lot more predictable in most circumstances than city driving, so as a problem to solve it's likely a little bit less complicated. What I wonder about is what it feels like as a passenger. I wonder if it would be more or less frightening than being a passenger when my 17 year old is driving.
I use adaptive cruise control a lot, where I rely on the car for keeping a safe distance.
I have a limited version of SuperCruise which means it operates hands-free on freeways but nowhere else. My wife's Equinox EV has the regular version, which operates on a lot of arterials near us and has more capabilities. The first time that the Equinox signaled, changed lanes to pass, signaled, then changed lanes back was shocking.
We moved to a small town and drive a lot more than we used to and I find that having those capabilities really helps relieve the stress.
I will say that I move to the center lane when going through a notorious set of curves on I-5 in Portland because my Bolt doesn't steer as smoothly as I'd like near the concrete barricades. I wanted SuperCruise because it has a fantastic safety record. There are lots of times it's not available but when it is, I have near-total confidence in it.
I took a Waymo that drove on an 'expressway' which had a speed limit of 40mph and it was definitely a different feeling. I did feel a bit scared, at 25mph it feels like a gentle theme park ride, at 40mph it's beyond that and feels dangerous.
There was a good overview on here a while ago about the challenges[1]. You need to plan longer in the future and your sensors need to reach further. It's also a much bigger challenge to collect sensor data as fewer diversions happen per mile (but those that do have higher stakes).
Roads that get used more collect more debris. They also break and require maintenance more often. That maintenance is exceptionally disruptive to the normal operation of the road.
Other drivers aren't your only challenge out there.
From the article “ Pickups and dropoffs will initially start at SFO’s Kiss & Fly area – a short AirTrain ride from the terminals – with the intention to explore other locations at the airport in the future.”
Hopefully Waymo does a better job than SF Uber drivers. I can't tell you how many times I've had drivers make a wrong turn trying to find their way to the pickup point.
There's the big sign there telling you to go to arrivals for drop-off. This is probably a stupid question but can Waymo cars interpret those temporary display signs and follow them? Would it?
It seems to handle the standardized ones (think "construction ahead, detour left") perfectly well from the rides I've taken, but there's all sorts of ways they could be 'cheating' on that.
I'll be honest, I think LAX's traffic is better than SFO's.
It feels like there's a lot less spaghetti at LAX, the shortcuts are reasonable, and you don't have separate international and domestic loops.
The Big Dig, for all the digs it rightfully got for taking forever and costing a shitton, actually does the job it's supposed to (mostly). I'm generally pleasantly surprised how few problems it had when I lived there.
I haven't driven through it much since it was completed - I moved out of Boston in 2005. As a now-just-visitor I find driving through the central artery tunnels even more stressful than on the old elevated 93, but that could just be a lack of familiarity thing -- I don't drive when I visit Boston, only if I'm flying in to Logan to head to the cape or something. It is, however, _really_ nice not having the freeway cutting through and the greenway is gorgeous.
I'll take JFK over LAX. The construction going on right now at JFK sucks, but LAX is comically bad. Just last week I was on a rental car shuttle at LAX and watched 3 separate groups of people at different terminals miss their flights because traffic just wasn't moving.
Maybe, but this approvals only allows them to go to the rental car center, which is quite far from the terminals. The passenger will need to take the air train to the terminals.
If Waymo can pull off airport pickups smoothly, it might shift how we think about edge entry to city traffic. Most cities still struggle with that "last mile" problem maybe self driving fits perfectly there.
One concern I have is how the user data collected by self driving cars will be handled. Companies like Waymo likely hold even more data than Uber. If such data is truly used in sensitive locations like airports, I hope there will be clear and transparent mechanisms.
Can you handle parking structures? I heard a lot of the autonomous cars were using 2D maps and couldn't handle multiple levels. Haha! This was just a year or two ago.
Google maps has been able to figure out parking structures for me recently. Not sure what technology is involved (gps isn't great for vertical) but it's clearly possible.
Google has been collecting data on building interiors for several years now. Not just parking structures. This data is currently used in streetview. Google's geospatial data is unequaled and maybe a bigger advantage than is readily visible.
Do they need a "map" of a parking tower though, just like how humans don't exactly need Google Maps inside of one? I feel like this is something self driving + vision (exit signs and arrows) can handle
There is viable public transit from SFO to the city. Both Caltrain and Bart have decent connections to the airport.
Practically, travelers with heavy check-in bags prefer taxis. Public transit can't be the answer for everything.
Without heavy check-in, transit is a great choice. Alas, some problems remain. Many residents consider BART unsafe. Frequency is pathetic for a major subway line. BART is unusually expensive ($14! - 6$ + 6$ + 2$??) and last mile connectivity is bad.
On the topic of cost, other airports that are on the transit corridor (LaGuardia NYC, Boston Logan, Ronald Reagan Washington) cost $3-4 for the trip. EWR & JFK are more expensive, but they're often faster than taking a taxi which allows them to compete on pricing.
Not sure if you have a recent side-by-side example with Uber, but this seems like it would have to happen if the demand is there. How else can you offer a quality product (i.e., car shows up in a reasonable amount of time) if you don't have enough cars to satisfy the demand? Pricing is the primary demand lever.
There's so much polarizing opinion on Tesla's offering and whether they'll get to Waymo's level sooner than later, but this seems like it's going to be or already is a huge issue for Waymo where they can't manufacture the vehicles fast enough to satisfy the demand as they expand both locally (because they capture more of the market) and into new geographies. Will they try and acquire a manufacturer? I don't think that's economically feasible for Waymo (Geely market cap is $25b, per Google snippet fwiw), and obviously being in the car business is different than autonomous, but I'm sure Google would bankroll a purchase if they thought it was the right growth strategy.
I guess Tesla, even if their autonomous is on par with Waymo tomorrow, also has to manufacture the fleet, but it seems extremely beneficial to have that capacity in house vs. relying on partners. Maybe I'm wrong and it's not that much of an advantage, but at first glance it would seem to be.
The CEO of Uber was quoted as saying Waymos complete more rides per day than 99% of Uber drivers. He didn't give a precise ratio but this makes me think that hundreds of Waymos can replace thousands of Uber drivers and their cars.
CMs like Magna have the flexibility to manufacture, at the low end, hundreds of vehicles, and at the high end thousands. I doubt Waymo will ever make their own vehicles. They are already working with Toyota on adapting Waymo technology to privately owned cars. That implies mass production. That would be a supply of vehicles that are probably simple to adapt to robotaxi use.
That's a crazy statistic and an interesting one for him to actually say out loud. Was that in the context of Uber partnering with Waymo in Austin? And thanks for the insight on the manufacturing side. Sounds like it might actually be to their advantage to use third parties because you can spread the demand around and since auto margins are not high the added cost for that benefit is minimal.
Private cars have a ridiculously low duty cycle. They mainly sit around waiting for their owners to use them. I suppose at some point in the future there might be a traffic jam of autonomous vehicles, but only if the providers are antisocial and don't coordinate ride destinations and routes.
@harmmonica I do. we prefer to use Waymos in SF but Uber has been a lot cheaper in the last six months or so regardless of time of day...
Also saw some Zoox self driving boxes on the las vegas strip last week but no one seemed to be using them.
Thanks for sharing, oliver. For anything local I've almost entirely switched and I guess I haven't been doing much price comparison between the two. One thing I have noticed here in LA, albeit only a couple of times, is that during rush hour the waiting time is significantly longer for Waymo. I've taken some Ubers because the wait for a Waymo has been way too long.
I think self-driving is what makes Waymo premium. A Waymo is a better experience because the Waymo is safer and smoother driver than any human could ever be.
Not sure why you're downvoted. I've tried Robotaxi a few times and has been great. They still have a safety driver these days and wait time is a big high though.
It will remain higher for a while. From reporting I have seen, they are close to maxing out their vehicles, and many people prefer it to other options, so are willing to pay a premium. As long as that is true, it's going to be priced as a premium product. It won't be until fleets grow significantly in size and/or another driverless taxi service enters the market that we will maybe start to see prices driven down closer to marginal cost of a ride.
-edit- multiple other comments apparently disagree with this. I'll defer to people who actually use them over the reporting. Odd that there is that disconnect though.
It's also higher right now because it is a novely. Plenty of people are booking it just to say they rode in a Waymo and take pictures. When that wears off they will have to start competing strictily on price and wait/ride time.
To be fair, Waymo claims to not record or transmit audio without you either manually engaging such (by requesting support), or a very unambiguous announcement (presumably when the car gets into some sort of emergency situation). And lying about that claim would probably run afoul of California's 2 party consent law. So still a step up in privacy versus having someone in the car listening in on your conversation.
That said, even if they were listening to you, there's a lot of things that are completely inconsequential from a perspective of an anonymous call center employee far away listening in on, that I probably wouldn't want to talk about in front of a taxi driver.
I know this is somewhat besides the point of the discussion, but.. many Ubers have recording devices inside the car too. The drivers have gotten savvy and protect themselves from false claims or even harassment.
This is just me, but maybe helps explain it. It's not that the presence of a driver is bothersome, but in the pre-Waymo world your interaction with the outside world starts when you step out the door of your house. Now the interaction with the outside world starts when you get to your destination and step out of the Waymo. I really enjoy the outside world, mind you. But it just feels easier to traverse my local area in solitude and with a consistent and comfortable vehicle, and non-erratic driving style.
I imagine how nice it could (will?) be when you can hop into a self-driving car for a longer ride or even a road trip. I think you'll feel like it's an extension of your living room vs. being in a car.
If you step back do you really think that's indicative of a mental disease? Does it make any difference to you that many times I'm taking a Waymo to go and hang out with friends? Not much of a stretch to say it's allowing me to socialize more because I don't have to worry about my meter running dry, or having one too many drinks to drive myself home, or being able to move around from area to area in comfort. And if you say "you can do that with an Uber too!" it's true! But does it really surprise you that someone would want a car that drives calmly, obeys all traffic laws and gives you a little downtime from the outside world pre or post the activity you'd been doing before stepping in the car? Does that really rise to the level of mental disease?
It seems like a huge catastrophizing stretch to get there based solely on preferring to be in a Waymo rather than a taxi or Uber.
Well on second reading your comment reads like an ad, given advertising isn't natural either, you can understand the confusion
But I was referring to the wanting the outside world to resemble your house and to have little interaction with humans. No, that's not normal, despite any sophistry or ad speak
The problem isn't when they don't talk and just drive, the problem is when it's late at night and the passenger is a woman who is inebriated. Not having a driver entirely makes that much harder.
Yeah they need scaling and competition before the prices get lower. As long as supply is saturated with demand and nobody else is on their level, there's little reason to lower prices.
Yeah, and just to add even though it's implied in your comment, there's plenty of reason to keep prices where they are independent of a desire to increase revenue. Customers will not wait forever for the car and so if the demand is high you have to keep the price high to discourage people from using it so wait times remain in check. Tricky tightrope they're going to be walking while they optimize the fleet size for local adoption and geographic expansion.
On other threads I've seen conflicting anecdata regarding pricing being higher or lower than an Uber ride. That's not too surprising since the supply and demand variables are going to be different for Waymo.
In my experience so far, Waymo costs about the same as an Uber when you take into account tipping, but takes longer (they're not yet doing freeways). With the addition of SFO to their zone, I can't imagine freeways are far behind, because getting from the city to SFO without using the freeways would be... a novelty.
That's not been my experience... 90% of the time when I check, Waymo is still a good 20-50% more expensive in SF, when comparing to a tip-included Uber or Lyft price.
I've used Waymo countless times in SF. It's typically 15% cheaper than an Uber/Lyft and trip time/wait are generally the same. I much prefer the Waymo.
They still have to compete with alternate modes of transportation such as buses, bikes, trains, e-scooter rentals, self-owned cars, Uber with human drivers.
If it would be "too much", then there's no reason why taxis (incl uber/lyft) wouldn't be too much today.
Direct competitors are uber and Lyft which they can undercut since they don’t pay drivers.
The people who want to take buses and trains will continue to do so although Waymo might sway some with their ease and if pricing is reasonable.
Bikes and e-scooters only get you so far. Last time I was in SF I didn’t see too many bikes but I saw a ton of e-scooters. Are you really taking an e-scooter further than a few blocks? And when it rains?
Self owned cars make sense for longer trips out of the city but parking is a pain and driving is stressful so this is an easy win for Waymo.
It’s cheaper now so they can take market share. And their cost will certainly be lower than Ubers so they can win the pricing battle. But long term monopoly gonna monopoly. Perfect pricing is a given with the wealth in SF and how many rides will be on a business CC.
from what I heard, the intention is to make it much more affordable than it is now. I don't remember the source right now but I did think it was a blog post or something like that.
I think if it's affordable then people will easily take that. instead of drinking and driving at night or other unsafe activities. if it's affordable then people can just take a waymo home and then back again to get their car when it's safe again.
Certainly they aim to make it affordable now in order to undercut Lyft and uber. Long term they will own the market and jack up prices as monopolies do.
L5 means the car can drive everywhere a human can. Waymo's refuse to drive outside of a constrained area, and occasionally stop to ask for assistance, so that makes them L4.
This whole autonomous driving levels kinda muddies the waters. Some would argue this isn't full L4 even. But it is a self driving car in the places it offers its services.
I think there's an implicit "where a decent human driver could drive safely" for L5, otherwise you get increasingly ridiculous scenarios like, "can Waymo drive safely in a whiteout blizzard?" or "can Waymo safely escape an erupting volcano??"
haven ridden in both a few times, yes, Waymo is head and shoulders better. It's smooth and I don't think I've ever seen any false alarms or behavior that made me feel unsafe in a Waymo, while I've had a few scary or annoying situations in the Teslas. I took a 6-minute robotaxi in drizzling weather where it parked in intersections twice because the cameras were obscured. Meanwhile Waymo can drive perfectly in heavy fog.
Both the Waymos and Teslas have that central display that shows you what the car sees (pedestrians, dogs, traffic cones, other cars, etc). The Waymo representation of the world reaches pretty far is is pretty much perfect from what I've seen. Meanwhile the Tesla one until recently had objects popping in and out.
Neither is perfect, of course; both will hesitate sometimes and creep along when (IMO) they should commit. But they're both still way better in that regard compared to the zoox autonomous cars I see in SF.
Tesla doesn't have a real robotaxi yet, they're still in the testing/prototyping phase where they need a safety driver or safety monitor in the car.
They might be close to a real robotaxi in some areas, but it's hard to say until they actually pull the trigger on removing any employees from the car.
Waymo cannot scale. So for most people it's irrelevant.
Tesla FSD makes driving 90% less taxing mentally. It does 99.9% of the driving perfectly. And its getting better. We are quickly approaching a situation where people who don't drive Teslas are like people who cut their grass with Sickle as compared to people who have driving lawn mowers
An interesting thing about this is that there are fewer than 1000 Waymos in the SF service area. I don't know today's total, but I'm pretty certain that there are fewer than 5000 Waymos in existence. Maybe as few as half that.
Some months ago Waymo claimed to be providing 250,000 rides per week. If the fleet size was 2500 at the time, that would be 100 rides per vehicle per week.
What’s special about the airport is that the City of San Francisco owns and regulates it (as opposed to the streets that are regulated by the state CPUC), and the Board of Supervisors previously were regulatory captured by taxi medallion owners and Teamsters union (https://missionlocal.org/2024/12/waymo-rolls-toward-san-fran...). Specifically, Aaron Peskin (BoS supervisor from 2001–2009, 2015–2025, and board president for the last 2 years) said, “Their entire M.O. is, ‘The state regulates us; we don’t have to work with you, we don’t have to partner with you.’ My response is: There are things we do control. Including where you charge your cars. And the airport. What I intend to do, is condition their deployment and use of the airport property on their meeting a number of conditions around meeting this city’s minimum standards for public safety and transit.” https://missionlocal.org/2023/11/waymo-rebuffed-by-sfo-sf-gu...
I’d say it puts a lot of Uber (and similar) drivers at risk because airport rides are a good source of income. Waymo undercutting them will reduce the amount of passengers available for pick up.
Not saying it’s a bad or good thing. Just that it has real world impact on people and the economy.
Usually you'd have to take the BART one stop then the waymo, which seems to be a common tourist attraction for fresh deplaners. Perhaps the airport was afraid without that step of friction, too many people would try this and cause a waymo-jam
They aren't the only autonomous vehicle for hire service. Zoox is operating in Vegas.
Even if they were the only one, it would be odd to classify autonomous rideshare as a distinct market given they compete directly with other vehicle for hire services where they have nothing close to monopoly-like power.
Inside SF, my experience is that Uber and Lyft are ~10-15% cheaper than Waymo, but that's before tipping. I don't have to tip a robot, so they work out to nearly identical prices.
The inherent problem there is the edges, most food delivery isn't the trip, it's the person getting out of the vehicle and putting it on your doorstep or going through the building. Zipline and their droneports for buildings seem to have the better solution, at least until waymo has some sort of legged robot that can bring the bag the last meter(s)
I think the frustration with tips is so prevalent that the advertising could just be "Skip the tip, simply walk to the street to pick up your order!"
Would work great in suburbs where a robot car could pull in front of home for a minute or two, your food will be bid to another customer if you don't pick it up in 5 minutes. maybe the little robots in NYC are better.
I would argue that the sidewalk robots are too hard to coordinate and not strong enough to hold up against crime, the solution is somewhat closer to my other comment below, a vehicle with maybe 4 or 8 food cells that can fill up at various locations then make its journey around the city. At that point the problem would be idle timeouts and how to handle disgruntled consumers that lost their window for pickup
Aren't the "first meters" also pretty problematic? Are Waymos going to double park in front of a restaurant waiting for someone to come out and put the right order in the right vehicle?
That's easier to do with training, and the business is usually more willing than a consumer as it increases their business. Anecdotally, see how many of them (at least locally to myself) have adopted the doordash/grubhub tablets in their kitchen ordering system. I imagine it would be a co-packing situation with lockers on wheels similar to the vehicle KFC uses in China: https://www.mashed.com/284555/the-futuristic-way-kfc-is-sell...
Nuro is an independent company from Uber, the latter just has a partnership with and some investment in the former. Uber has similar relationships with more than half the industry at this point.
My anecdotal evidence also has so many incorrect orders that I'm a wee bit less optimistic than you about restaurant-side human handling of the first meters. :)
I have a relative in Texas who is looking into leasing a drone to operate for food delivery. Apparently, that's already a thing there? If we could get food/small packages delivered to our building's roof instead of the front door, it would be a huge win for everyone in the building.
The longer the route, the harder it is for the food to stay fresh and warm/cold/frozen. It's a trade-off between efficiency, price, and customer satisfaction.
Self-serve ordering terminals already often ask for tips. Presumably to be legal they're being paid to the kitchen staff, but I think sticking to "tips are for workers who have to pretend to like me" is a pretty firm boundary to stick to.
(Also, arbitrarily reclassifying things as tips is hard, because legally 100% of tip revenue has to go to workers, not management, and certainly not the company's investors or coffers).
Tax-free tips paid to robots go to the hardworking AI engineers -> AI engineers voluntarily donate part of their tips to a 501(c)(3) that helps support struggling venture capitalists.
Something like that. We'll work it out the details once the right PAC donations are in place.
Cause what this country needs is to automate away even the gig economy jobs that are out there. Let's keep making a few people rich and screw all the normal people out there.
Why the downvotes? That jobs will be lost is fact. Does this represent an increase in wealth concentration? Obviously. Is that a net bad? I don't know, let's discuss instead of silencing people.
Okay what're the odds on how long it is until there's a stray Waymo on the tarmac. Hopefully with enough warning to divert any planes about to land on it.
Who at Waymo can I speak to about using Waymo’s as an affordable housing solution? I work in commercial real estate and have a handful of affordable projects I am involved in and believe this to be a very interesting solution no one is talking about.
Waymo are toast, Tesla will out scale them in months if not sooner. They can't compete on costs, 100k plus for an ugly Waymo vs < 40k for Tesla model Y or cyber cab.
This sentence was a bit cute: "Waymo has received our pilot permit allowing for commercial operations at San Francisco International Airport." Yeah, that kind of pilot.
I really had to read through it twice to make sure they were just talking about car taxis picking up travelers, rather than some kind of prototype pilotless commuter helicopter or something.
That was my first interpretation, and I was very surprised and kind of afraid. Glad to know they aren't trying for autonomous flight yet.
I have zero expertise for my claim, but I feel like autonomous flight is easier than autonomous driving.
The hard part of automated driving is dealing with all the ground clutter that planes serenely fly over. If pedestrians could charge out in front of a 777 going 650 mph at 34,000 feet... well... we'd be living in pretty different world! And in that world, flying would be much more difficult. Not just for computers but for humans too.
Flying is obviously much harder than driving, but it's a sort of harder that is generally more amenable to automation, though I still think pilots are a good idea because when it goes wrong it goes wrong much worse.
Flying is almost always easier than driving. landing is hard. Bad weather is hard. But just flying - human pilots have napped many times over the years and it only rarely is an issue. Airplanes with primitive autopilot are very good.
Yeah, a primitive autopilot in a plane just needs an altimeter and compass, but a AoA sensor, speedometer, fuel level sensor, and pitch sensor help to detect unsafe conditions like runaway pitch, stalling, overspeed, low fuel, etc. Each of those sensors is providing a simple 1-dimensional data point. Redundancy is relatively inexpensive.
Automatic lane keeping in a car requires cameras that software needs to then analyze to find the lines in the road in real time. But if you want a "set it and read a book for an hour", then you have to respond to other traffic. No longer just some simple PID controllers, the software now needs to plan and execute based on surrounding traffic.
Yep. 0ft-1000ft AGL Takeoff, Climb, Approach, and Landing are the tough bits. The rest (Cruise) is very low demand and much easier than driving.
Taxiing is probably harder to automate than the rest. But you could have pilots on hand to taxi to the runway, and take a shuttle to the other end and hop on a just landed plane to taxi to the gate. Or you could use tugs for ground movement.
I'm not convinced - in a commercial airport taxiways are controlled by a ground control systems, not just pilots looking out the window. If the only airplanes around are also equips with the self taxi system they just report position to the central control and that tells them when to go. There needs to be emergency overrides for when that system fails, or a small plane without it is around, but that can be handled by stopping everyone else in the area until the hazard is gone.
Time will tell...
Do you have a source for that? As to my knowledge advanced systems (such as lights on the twy directing you) are only present at very few airports. Recent incidents even happened due to RWY incursion without a ground controller noticing under bad visibility. So we are at a level where your runway is not even protected accordingly, let alone your 50+ taxiways.
There's also all of the service vehicles when you get closer to the gates. The likely damage from an incident during taxiing is much less than during take off or landing, but I think the risk of having an incident is higher and the situation is trickier to manage. And it's super doable to have a pilot come on to manage that, and drop off after the hard part; you couldn't reasonably have pilots do a takeoff and then jetpack over to an arriving plane to do the landing, but it wouldn't be unreasonable for ground moves... similar to canal pilots taking ships through canals.
>human pilots have napped many times over the
months?! :)
"The [German pilots'] union said it had carried out a survey of more than 900 pilots in recent weeks, which found that 93% of them admitted to napping during a flight in the past few months."
-The Guardian, "Almost all German pilots admit to napping during flights in union survey"; 2025-09-10
Years as since humans have flown planes stable enough not to need constant attention. On a calm day you don't need autopilot, just set your trims correctly and some airplanes will hold course well enough for a short nap - though of course this is more likely to result in a crash (which likely has happened, though it is hard to guess why a plane crashed beyond pilot error)
Yeah, but a 1% angle over a long period of time intersects with the ground, and I wouldn't want to trust your alarm clock with 200 lives.
The pilots would get an extra "Woop! Woop! Terrain! Terrain! Pull up! Pull up!" redundant alarm clock, although it's terrifying to rely on it.
I wouldn't want to either, and even the pilots who have done it claim accident. It has worked out for a lot.
Not to mention that almost all civilian planes in the US are required to broadcast a bunch of details that include their coordinates and altitude on a public channel (ADS-B). It's the kind of automated collision avoidance input that you'd probably dream of as a self-driving system engineer. Basically the only thing you'd need to avoid via more complex systems is the odd military traffic, small craft at low altitudes, and birds.
In the abstract yes but in practice the economic (ratio of cost of pilot to pax miles) and safety context of aviation mean fully autonomous flying has to be extremely robust before it has actual utility in industry.
In practice, you're also currently very reliant on infrastructure that is definitely not as solid as you want (eg: ILS and GPS can be interfered with quite nastily).
ILS being under maintenance and unavailable for certain runways is also far from unusual.
Commercial pilots are also extremely good at dealing with edge cases you wouldn't design an autonomous system for no matter how solid the infrastructure, like deciding the Hudson river is a good place to ditch
And their cost relative to other operational costs is so low there isn't even any pushback on regulations regarding there being two of them.
Pilot cost isn't low. The airline industry is very much looking for ways to reduce crews, whether that's going to single pilot operations (long term) or reduced crew operations (short term).
RCO is very much "pushback on regulations".
On the happy path, yes. Though I don’t think takeoff is automated yet.
Currently we rely very much on the problem solving abilities of human pilots to deal with troublesome situations. Autopilot will disengage in many scenarios.
I'm pretty sure drones can already take off on their own. Taking off is a lot easier than landing, and planes have auto-landing tech already.
Drones (both autonomous and remote piloted) have much higher mishap rates than crewed aircraft. Taking off is "easy" until something goes wrong, like a mechanical failure or runway incursion. It's impossible to anticipate and explicitly code for every possible failure mode, so developing autonomous flight control systems that would be safe enough for commercial passenger flights is extremely challenging.
Category IIIC ILS (full auto-land) does exist but requires special equipment for both the aircraft and airport. Human pilots have to actively monitor the system and take back control if anything goes wrong (which does happen).
Garmin also has the Autonomí auto-land system for certain general aviation aircraft which can attempt to land at the closest suitable airport. But this is only used for single pilot operation in case the pilot becomes incapacitated. It isn't suitable for regular flights.
Consider that drones may fail more because failure is an acceptable outcome for drones.
OK, I've considered that and determined it to be mostly wrong. While drone failure is an acceptable outcome, current technology still doesn't allow drones to be as safe as equivalent crewed aircraft across the full range of flight operations. Maybe in 50 years we'll get there.
Sure current tech does not allow us to safely automate flights. What I wanted to get at is that tech that doesn't need developing does not get built. Looking at drone failures does not tell us the max safety they would reach if we focused on that.
Obviously a droneliner would look very different from the jets that are common today.
Takeoff at a commercial airport is a very challenging and potentially dangerous situation. There’s way more margin to abort a landing than a takeoff.
OTOH takeoff and landing could in theory be operated by people on the ground, flying simulator style.
I still believe that having an actual pilot inside the plane that care for his own life is not a bad idea vs someone remote feeling a bit disconnected with the reality of a crash.
Remote piloting is how the military operates certain drones like the MQ-1 Predator. The mishap rate is very high relative to crewed aircraft due to network lag and sensor issues. The military is willing to accept some level of equipment loss in order to accomplish their mission but this would never be allowed for commercial airliners.
The pilot’s self-preservation instincts aren’t the most important reason to have them onboard. It’s that any loss of communication between the ground and the airplane at any point during either procedure would turn it into an uncontrolled cruise missile.
I am not sure why you were down voted. The original meaning of the word pilot is someone who comes aboard a ship for "the last mile" - getting in and out of the harbor and what you are talking about is kinda like that - a person associated with the airport rather than airplane to guide the planes in and out - perhaps using more reliable local communication technology vs what is used to control drones half way around the world.
I have no idea if that works but I thought you were making a good contribution to the conversation by proposing a potential solution to the exact problem everyone is talking about.
It's kind of funny how you can both be right.
Drones crash on takeoff all the time. Worth noting that drones are more than just quadcopters and serious drones are often winged aircraft.
It's the failed takeoffs that lead more often to jets leaving the run way and crashing into buildings or trees.
0d (parked) - null program easy
1d (train) - easy. just one lever
2d (car) - hard. super hard. why is it so crowded? who thought this was a good idea? you let teenagers do this?
2.5d (plane at takeoff or landing) - almost as hard as car. fewer pedestrians.
3d (plane flying) - easy even with all those extra levers
I'm not actually sure how hard landing is. Most airports that support autonomous landings do it by having ILS antennae that guide the airplane to within tens of feet of the runway, at which point the airplane switches to radar for altitude.
Automatic landings started in 1964. I think that it seems hard mostly because of how tightly regulated aviation is - modern technology could probably make things a lot better if people were more receptive to the idea of heavy automated aircraft over populated areas.
landing is easy. the hard part is landing with 20mph cross winds and one engine out (or other mechanical failures). we've had auto-land that is 99% reliable for a while now, but you need to get to 6 9s before you have a system safe enough to replace pilots
I think that as long as the autopilot is able to fly in a crosswind or with an engine failure, it can probably land with one. Autopilots are already able to do these things.
I doubt anyone has tested this in depth, but I'm not sure there are too many configurations of airplane these days where a human can safely land it and a computer can't. Maybe if a big chunk of wing or control surfaces were totally gone, but even a human pilot isn't getting 99% reliability in a situation like that.
In any case, I don't think that the first candidates for automation are gonna be passenger flights. It will probably be small cargo planes first - Cessna Caravans and other turboprop aircraft where the cost of paying pilots is roughly similar to the price of fuel.
1d, variant (tram) - hard, who thought it was a good idea to send rail-bound vehicles and steerable vehicles down the same road?
3d, variant (orbital) - super hard, so hard that trajectory pre-calculations has to be performed
My layman's understanding is that we've been doing it already for decades with expert system "AI", so likely much easier than navigating streets with other people.
Not a pilot myself but it seems that a large part of the danger with flying is that when something goes wrong you are much more likely to have a high speed crash. Cars don't even usually travel at speeds that planes crash at.
When everything is working correctly, no other pilots have emergencies, and no temporary restrictions are in place, and there are no clouds in the sky. Then yes, it /could/ be easier, but almost always it never actually is.
There's a reason the majority of accidents occur during take off and landing.
Spend some time listening: https://www.liveatc.net/search/?icao=ksfo
It depends a bit on your safety standards. There are already autonomous flying things delivering blood and blowing up oil depots where it doesn't matter so much if stuff goes wrong, but to be an airline pilot you have to know how to deal with a huge range of emergencies and systems packing up.
With a car if the engine fails you just pull over. With an airliner it's not so simple. As a result the training for a pilot is much longer than for a bus driver say.
I also feel like the demand is way, way lower. A pilot can't be that large a % of the cost of a flight. Maybe if we lived in the jetsons era.
The problem is actually safety. As automated systems get better, the pilot is left with not much to do, and has to maintain vigilance while being really really bored. It is almost better to have fewer automated systems and give the pilot more things to do during the flight so it is easier to keep them paying attention, or all automated with no human pilot to mess things up.
Depends on the size of the plane, really. One of the reasons a few companies were investing in fully autonomous air taxis is because the math on a small piloted aircraft wasn't realistic for a low enough price point to be competitive.
Navigation might be easier. The battery and safety tech isn't there yet to make it practical.
When shit goes wrong for a car (such as Waymo) you just stop. Now, that's not trivial, but it's also not very difficult, I expect most of it can happen even if the Waymo hardware itself were suddenly destroyed, rolling along is the hard bit so not doing that isn't too difficult. Everybody aboard can just leave when it stops moving.
In contrast when shit goes wrong for a plane we've got a big problem. Just stopping will definitely kill everybody, even from a modest altitude at a very low speed suddenly plummeting to the ground will straight up kill you. So, we want to land, albeit maybe we have to "crash land" destroying the vehicle to perhaps save its occupants.
You can buy (and indeed to some extent you can even retro-fit) emergency auto-land for small planes. Once engaged, or if set to do so automatically upon pilot failure the plane will figure out where it is (using GPS), pick the emergency radio frequency and announce the problem and its intended solution (I am a machine. My human pilot is incapacitated. I intend to fly to X location and land there. I am not listening to you and cannot understand you) and then it will fly to a chosen place and attempt to properly land the aeroplane, broadcast on radio that this airfield is now closed (this aeroplane is parked on the landing strip so you can't use it!) and then switch off.
Maybe the pilot is still alive and human medics can rush them to hospital. Otherwise maybe there are passengers who have been saved. In any case at least the aeroplane is now on the ground where humans can easily take over e.g. moving the plane so the airfield can re-open.
In a pinch, a car can just put on its hazards and pull over
That “just” is doing some heavy lifting! The car still has to deal with all the normal hazards of the road while pulling over, plus the hazards it is itself creating by acting abnormally.
Could also be a big challenge if you have dozens or hundreds of autonomous cars in the area that need manual intervention to get them out (plus the people who get stuck there)
Is that situation somehow less difficult with aircraft?
Well if we're being picky, technically the car itself doesn't have to deal with the hazards it has created, rather everyone else does.
The point is you can't just "stop" a plane and wait for someone to figure things out (https://support.google.com/waymo/answer/9449023?hl=en). Whatever the difficulties in dealing with an abnormal situation in a car, it is strictly much more difficult to deal with them in a vehicle constantly fighting the homicidal urge to fall out of the sky.
They each have their own unique issues. Being in a pinch is not universally harder for a plane.
Also constant urge to fall out of the sky is a helicopter. A plane generally wants to glide.
Or a fighter jet. But definitely not a passenger plane; those things settle into a phugoid cycle.
Ok cool, so a plane merely needs to continue at highway speed, in the face of any difficulty, be it mechanical, electrical, software, weather... Like the movie Speed.
Don’t have a ref but heard that it’s been safe for quite a while but they keep the pilots around due to consumer fear rather than actual improved performance. Curious if anyone can confirm.
No. Airliners can't even take off on their own yet, and are only allowed to auto-land with zero visibility at a few dozen airports when the pilots, plane, and runway are all current/recently checked.
Look up the Airbus ATTOL project's first automated takeoff a few years ago.
Also, there's virtually no automation when it comes to interacting with ATC.
An airplane will take off when it is properly configured and it hits a certain speed. It's simple aerodynamics/physics. Pilots are there to react to failures and unexpected events.
There's a bit more to it since you do need to do last bit of configuration (pull up the nose) just as you hit the target speed. But yeah, automatic take-off is quite a bit easier than automatic rejection of take-off.
Even manually pulling up the nose once you reach Vr isn't necessary if you just trim for a little extra nose-up. It'll eventually get off the ground with just enough speed.
There's no lack of online arguments about whether or not Vr is "real" or should exist.
I just followed what my CFI and Cessna's manual for the C172 said (which iirc was giving input to rotate at 55kts).
Sure. It'll also land if you don't care about anyone surviving.
And the air is within acceptable temperature and pressure ranges. I assume configuration takes weight into account as well.
Yeah temperature, wind, altitude, weight, runway slope all matter, and then there needs to be enough spare space for the aircraft to successfully take off even with engine failure in the worst possible moment. Then there's the question of fuel consumption too. Takeoff power isn't typically configured to get the aircraft off the ground as fast as possible, but to minimize fuel consumption, while still leaving enough margin in case of engine failure.
It wouldn't be that hard to fully automate a flight from gate to gate when everything works perfectly. But the various failure modes, human error like airport vehicles entering active runway, all that requires human backup. Self-driving car can just stop to the side of the road and turn on emergency lights if its engine fails, with a plane things get much more complicated.
One of the hardest parts is just getting radio comms right. ICAO phraseology is supposedly standardized but when anything unusual happens then things get messy, especially if there are multiple aircraft involved.
Cars can drive around without needing to talk to other cars or controllers.
And you don't need rudder input or any aileron input because of crosswind, and other bits that falls into "technically correct but not particularly relevant" territory.
It's fun to see/feel planes do stuff "on their own" (eg making them oscillate, or level on their own, or feeling ground effect, or even your own wake on steep turns) but it's not something you'd want to rely on (maybe with the exception of ground effect on short field takeoffs, but I digress).
> Also, there's virtually no automation when it comes to interacting with ATC.
Check out the Cirrus Autoland feature in their aircraft. They are all small personal aircraft, but the tech is pretty cool. Will talk to ATC and fully auto-land for you in the event of an emergency where the pilot is incapacitated.
"Talking to ATC" is a bit of a huge ask. The system basically just hops on 121.5 (and maybe the nearest/local unicom/tower frequency) and start an automated callout with its intentions that it will be doing. It operates on the assumption that all other airspace users will hear the radio calls and stay clear of the emergency aircraft.
I'm aware of it, though I've never flown a Cirrus. But AFAIK, it announces what it's doing. It's not communicating.
If you can design the product and environment to fit automation, then automation can be quick and effective.
The less you can change about the product and environment, then automation run slower and less effectively.
Air liner operations could be automated, but the minimum equipment list would be more stringent, the destination airport would not be able to take any equipment out of service for maintenance, visibility minimums would increase, takeoff and landing operations would require more slack time.
Besides all of that, the owner of the airplane would still want to have some crew on board.
In short, it's not worth it yet.
===
There is also the paradox of automation: Automation generally makes the hard parts harder and the easy parts easier.
The current goal of autonomy for airliners is single-pilot operation more than full autonomy.
It's very cool stuff, technology wise, with potentially significant redesigns of cockpits, etc.
But the main thing is the plane basically needs to be able to operate just about entirely autonomously (especially during critical flight phases) in case the pilot is incapacitated.
In theory, once SPO is solved, autonomy is almost solved.
I'm skeptical that SPO will be allowed for commercial airliners in our lifetimes. Pilot workloads are fairly low during most routine flights. But when an emergency occurs then the workload suddenly gets extremely high, to the extent that even two pilots are sometimes overwhelmed. This isn't a problem that current automation technology can solve. There are an infinite number of possible emergency scenarios and engineers can't possibly code for and test every one.
Cargo flights over oceans and (mostly) unpopulated areas might be a valid use case for SPO. Cargo pilots have always been considered somewhat expendable.
I watched video about incident where plane was really lucky that there was a pilot riding along in the jump seat when engine went out. The pilots were wrestling the plane and the extra guy was able to debug the real problem. Maybe it was figuring out which engine was on fire and shutting it off.
I'm maybe less skeptical than you but still not super positive.
At the very least, I'd say it's at least two clean-sheet designs away (which I'd guesstimate at 30 years).
I'm a bit partial to it because I did a brief stint in the Airbus realm. Autonomy for airliners is an interesting set of challenges.
What would ever make you think that?
In an automotive setting you can almost always safely decelerate to a full-stop, put on hazards and call it a fail. Good luck trying that in an aircraft over urban areas.
"Autopilot" already exists when it comes to flying.
Sure but it's not autonomous in the sense of Waymo (ie, driverless)
Landing can be: https://en.m.wikipedia.org/wiki/Autoland
In fact, it's pretty routine. Don't have the source at hand, but somewhere around 1% of all landings (at airports with ILS) are autolands.
I think it was Boeing that even requires at least 1 autoland per plane every 30 days or so.
You can find videos of this on YouTube. Completely hands-off.
Most carriers have a rule that on clear days you always hand fly the landing.
This is a competence you do not want to lose.
It's also the case that you can have a whole approach setup in your flight computer and at the last minute the controller gives you a runway change. You could drop your head down and start typing a bunch info the FMC but you're generally better off just disabling auto pilot and manually making the adjustment.
I'm curious, what is harder to implement: autoland for airplanes, or autoland for rockets (spaceX)?
I don't know if these are comparable.
But two interesting data points from the Wikipedia article I linked are that the first aircraft certification for ILS Cat III was in 1968, and Cat IIIB in 1975.
And IIRC by the 1980s, autoland was already a pretty common feature.
Yes but it should have been obvious that in the context of Waymo + SFO, the implication was autonomous flying of commercial airlines.
Yes, but autopilot usually just keeps the plane flying in a straight line at some specified altitude, which have been around since 1912. It isn't full self-flying (although we definitely have drones that can fly themselves already, so that tech already exists).
That's an oversimplification of autopilot systems. They can follow flight routes, avoid traffic (TCAS), even auto land to name a few.
Auto-landers are not simply classified with autopilots. An autoland system is an advanced function that is part of a modern aircraft's overall autopilot capabilities. A basic autopilot can control an aircraft's attitude and heading, but an autoland system can automatically execute the full landing procedure.
Mine as well, and I was crossing SFO off the list of airports I'd connect through.
[dead]
Cool, I wonder if this means they will finally start letting foreign visitors also use the app. I'm an American living abroad now for many years, and I was initially super excited to try Waymo in LA and SF this summer when I visited with my family. Unfortunately they only make the iPhone app available via the US app store, and while I actually have a US credit card that I could have in theory used to make the switch, Apple makes it an absurd pain to change your region as they require you to both a) cancel any existing subscription AND b) wait until they all expire. Most tourists have it worse as they have no option to even switch in theory.
Huh, as a Brit, I was able to use Waymo just fine on this summer on my Android device.
See also: garden-path sentence (https://en.wikipedia.org/wiki/Garden-path_sentence)
The flying kind is a license, not a permit.
There are driver's licenses and learner's permits. This could be the flying equivalent.
Honestly I think the title should be edited. The first time I scrolled past it I had the same obvious interpretation.
This is actually why you have "Naval aviators". To maritime people a pilot means something else. https://en.wikipedia.org/wiki/Maritime_pilot
lol deniable demand-gauging :)
As a European, I can’t help but feel a bit sad that we’re missing out on the driverless side of things. It seems like most of the meaningful deployments are happening in the US (Waymo, Cruise).
I’d really like to see either a Waymo competitor emerge in Europe, or even Waymo themselves operating here. The regulatory environment is obviously more complex, but it’d be great if we didn’t end up years behind on something this transformative.
Cars of any sort, self-driving or otherwise, do not solve traffic any more than Uber does because you need to have enough of them to get everyone to and from work at basically the same time. Trains are the only way to address traffic. Trains are self-driving. Europe already has the better self-driving system. It's just boring because self-driving is much easier when you build the road to support it instead of removing all constraints and adding GPUs, lidar sensors, cameras and an army of fall-back operators in overseas call centers.
If trips that require a car are prohibitively expensive (in money, time or convenience) without owning a car, more people will own a car. Once you own a car, it's often much easier to use it for trips that you would otherwise do without a car.
Reducing the (perceived) need to buy a car, e.g. by making it easy, cheap and reliable to get from A to B using a self driving car service, will reduce the number of people who own a car and thus the number of car trips.
Taxi services can potentially complement public service by filling in the gaps: last-mile connections (home to train station) and backup service late at night when transit runs less frequently or not at all.
There's a risk that robotaxis could become too cheap and people use them for point-to-point transportation because it's faster. This could be mitigated through taxes on robotaxis (with incentives to connect people to transit) and/or car usage in general, or maybe using robo-buses to provide a middle ground between personal convenience and system efficiency.
>Reducing the (perceived) need to buy a car, e.g. by making it easy, cheap and reliable to get from A to B using a self driving car service
But this assumes the need for a car, but cars are one form of transport. A more wholistic look at transportation with be “Reducing the (perceived) need to buy a car, e.g. by making it easy, cheap and reliable to get from A to B.” If you have more services within walking distance, it reduces your need for a car. If there is lots of bike infrastructure, it reduces your need for a car. If there are reliable frequent trains, it reduces your need for a car. If there are reliable frequent bus services and bus lanes to get around traffic, it reduces your need for a car.
On the other hand, if there are more cars then you need, at minimum if we imagine self-driving cars, more road capacity. But realistically more roads and more parking. More space for roads is less space for the actual places people want to go, pushing those things farther apart. Being farther apart reduces the number of places you can get to by walking or biking, which means you are more likely to need a car, which means more cars, which means more roads, which means less space for the actual places people want to go, repeat. Cars are basically the worst option in terms of infrastructure cost, land usage per person, personal cost to use/operate, deaths and injuries, etc.
This assumes the occasional need for a car, because even with the best public transit etc. in the world, there will be cases where not using a car is impractical.
If you need a car on at least a weekly basis, you're probably going to have your own car either way (unless the self driving car services are really good and cheap).
But even if all everyday trips don't require a car, it's very likely there will be some exceptions. And those can make or break this. If getting a car for that occasion requires hours of overhead (e.g. getting to a pick up/drop off point), is sufficiently inflexible (cars not reliably available on short notice), or prohibitively expensive (e.g. per-km charges on car sharing cars that make a couple longer day trips per year more expensive than just getting a cheap car), people who otherwise could do without a car will consider getting one.
OTOH, if the alternative is really good, people who occasionally need a car might use a service rather than owning a car, which means usage-based cost i.e. a much bigger incentive to pick alternatives. If they have been pushed to own a car, the fixed costs are a sunk cost and the marginal cost of taking car can easily be cheaper than public transit.
Sadly the evidence for Uber like services is that they take journeys away from public transport rather than encouraging its use.
The main effect of making the car more comfortable, in this case by removing the controls, is to encourage (subsidise) people to spend longer in the car.
So people will be willing to drive further for cheaper rent, or the self driving car might add a couple extra miles to park somewhere cheaper, so overall congestion would get worse.
Self driving cars could work with trains to do the desired location to the station bit that has always been a bit awkward.
Trains are all very well but they've been around nearly 200 years and have yet to bring on a traffic free utopia.
Can you imagine how much traffic there would be if NYC didn't have the MTA? The principle of induced demand tells us that as long as there are roads they will have roughly constant traffic because people are willing to spend some roughly constant amount of time getting to and from destinations by road each day. More roads speeds up everyone's commute which brings in more drivers, which brings traffic right back to the baseline terribleness.
The question is how shitty it would be if they also had everyone on them who's currently on public transit.
So basically, it is a traffic-free panacea for everyone who chooses to use it. It's not a goal of trains to eliminate traffic for everyone who insists on driving.
https://www.tomtom.com/newsroom/explainers-and-insights/indu...
The induced demand argument works for trains too. If NYC didn't have MTA (no subway, no LIRR, no MNR) then the population of NYC would probably be 1% of what it currently is. Building more train tracks and having better train services also encourages more people to move to NYC so that these new train services become more utilized.
Neither roads or train tracks solve the traffic problem.
Train density is high enough that you might actually be able to build enough tracks to keep up with demand. Tokyo has just about kept up with growth by building trains, and (unlike cars in NYC) the trains don't have to dominate the city to do that.
Yep, this is a good point. There are appropriate technologies for each situation. It's not a winner-takes-all contest.
For another example, can you imagine trains replacing school buses in a large, rural school district? Sometimes (not always), buses are better than trains.
Any one part would have the about same amount of traffic it does now. It would just sprawl out bigger across adjacent counties and the highest density parts would be lower density.
See also: LA
The tubes were shut down due to a transit workers strike recently in London. Here's what the streets looked like:
https://www.reddit.com/media?url=https%3A%2F%2Fi.redd.it%2F2...
Now imagine if all those commuters were in cars.
This is what bikes and busses are for, or just walking because the metro system is comprehensive enough you are at most four blocks away from a station.
> have yet to bring on a traffic free utopia
This is a silly expectation to have. As long as there are roads for cars people will put cars on them.
Trains solve traffic for the people who get on them, not for drivers. The more people taking the train, the fewer people impacted by the traffic.
You could maybe have something like Zermatt Switzerland which is car free but you can get around in human driven golf cart like taxis. It's pretty pleasant but expensive. If the carts were self driving it could be cheaper.
(Zermatt pics https://www.traveladventuregurus.com/zermatt)
Zermatt is fundamentally a pedestrian town. There are a limited number of permits for electric vehicles available for companies that have an objective need for a vehicle. That limited availability makes the electric taxis expensive.
The total number of permits seems to be around 500 in a town of 5k permanent residents. And the population grows to 30k or 40k during the peak tourist season.
Yeah that approach can't be scaled to cities. Folks go there to chill or do alpinism, not live their lives and work. Otherwise those narrow steep streets would have very quickly rush hours and traffic jams, its really not a place designed for any traffic apart from walking.
One day, cheap automated electric self driving taxis will cover cities, thats unavoidable I think, but we are not there yet.
Cities are experimenting with traffic free areas like Barcelona's superblocks. You could imagine something like that but with cheap automated electric self driving taxis added. I agree we are not there - Waymo basically just substitute normal taxis.
They help to remove some congestion in the Netherlands. That’s my everyday experience. Traffic would be way worse otherwise
> Trains are all very well but they've been around nearly 200 years and have yet to bring on a traffic free utopia.
Cars will always have a purpose. But if you go to somewhere like The Netherlands, they are much less relied upon - it's more about delivery vans than getting individuals to places.
Visit Tokyo and tell me they haven't brought about a traffic free Utopia
Human beings naturally take advantage of new conveniences.
If public transportation just encourages people to move to the suburbs and commute in every day you've actually just displaced the problem.
I think self-driving cars can still be beneficial even if they don’t help with traffic problems. They shouldn’t require so much parking in desirable areas (a separate problem cars cause), for example, and they could have a big impact on the lives of some disabled people.
Trains will fairly unreliably take you from one place that is not your home, to another place, which is not where you want to go, at a time that is probably not exactly when you wanted to arrive. Freedom of movement is incredibly important, and trains are very rigid in this aspect.
Well That’s certainly not been my experience when visiting Europe. In fact, it many cases it’s been the opposite - having a car would have been restrictive in any major city and a source of friction.
> having a car would have been restrictive in any major city and a source of friction.
Would a Waymo that you don't have to store, park, fuel, or maintain have been restrictive?
Well to the extent it draws people from public transit, yes because traffic makes being a pedestrian more unpleasant and waymos still are traffic. And increased traffic adds friction to crossing streets and they park obnoxiously, among other things.
So yes, they would be obnoxious at any significant quantity and also not really help with getting across the city since transit is pretty good
Human driven vehicles are a menace: dangerous, loud, dirty. Self-driving vehicles are entirely different: safe, quiet, no tailpipe emissions.
I'd easily take extra self-driving vehicles if it reduced human driven ones.
Well yes if we’re arbitrary limiting our choice to car based transportation that makes sense for mild climate cities. But why are we insisting on cars being the backbone?
No limits. Each option should be evaluated on its merits.
My contention is that in US cities the high cost of existing rail makes it uncompetitive for most uses, and there is no justification for building new rail.
Maybe not a greenfield project, but rail lines like the NEC could benefit a lot from relatively cheap fixes: removing sharp curves, improving scheduling operations, etc. We just need to get the flywheel going on this in the US
I like Waymo a lot, but the USA desperately needs both transport modes. Don’t think it’s an either/or.
First, I assume that "NEC" means North East Corridor which has a "high speed" train on Boston-NYC-WashingtonDC. Second, "relatively cheap fixes: removing sharp curves": You lost me here. That train must be about 20 years old now. If this was so cheap and easy, why not already done?
Hell if I know why it hasn’t already been done. All I’m saying is that the route slows down because of some sharper turns in some areas, and fixing it would be easier than making completely new lines/stations. I’m sure it would be much more expensive than similar projects elsewhere in the world
Cars driving at high speed over normal asphalt also generate a lot of tire noies and particulate pollution, even if they are electric cars. I found this video pretty interesting - some cities are experimenting with different road surfaces to reduce noise
https://m.youtube.com/watch?v=CTV-wwszGw8
id still like to have some human drivers around, to call 911 when i get stuck under the automated car
Trains are great when going to tourist attractions, especially in the center of old cities.
When you live and work in a city, they're much less practical.
This is quite the "I have never lived anywhere else other than North America" take.
Rail and other public transport in pretty much everywhere in the world are designed to serve commute first, tourist stuff second or third.
Public transport isn't just having some trains, or having only trains between major cities. It is designing whole commute routes from various urban and suburban areas to workplace. There needs to be regional and suburban links that arrive to metro and tram stations. Metro and tram have to operate very frequently to handle commuters. The frequency of the trains should adapt to the commuters in the morning and evening. They need to be convenient, clean and safe too.
Cities around the world are also much better balanced than NA ones. The workplaces and living areas are almost always mixed rather than having a "downtown" area where every office worker travels to. My area has many buildings with a supermarket, apartments and small offices in the same building. There are two car factories in the city next to one of the biggest urban parks.
I'm a European who has emigrated to the US, and knows both sides pretty well.
I agree that European trains work very well for commuting to and from the center of big cities. That's where the jobs and tourist attractions are.
But to go between arbitrary places A and B is usually quite painful. Often the best way is to go to the center, and then from there to your destination.
When I moved to the US and got a car, it was an unreal feeling! I could quickly travel anywhere at anytime!! Practically it felt like my comfortable travel radius increased from 10km to 50km.
Is that why the trains and trams are crowded around commute? Because people find them impractical?
> Trains are great when going to tourist attractions, especially in the center of old cities. When you live and work in a city, they're much less practical.
This is the most "tell me you live in America without telling me you live in America" thing I've seen in a long time...
America basically the only place in the world where in its cities, trains and other public transport aren't a major part of people's lives. In other places (Seoul, Tokyo, many European cities, etc.), even people who own a car will sometimes commute via train due to the convenience.
Come and live in Switzerland for a year and learn something.
Is this a serious comment lol
In effective countries trains run frequently enough that you don't need to consult a schedule and are less prone to unexpected delays than cars. Yes, they can't provide door-to-door service; like it or not, everyone travelling door-to-door in their private mobile living room during the rush hour is impossible if you want cities dense enough to be liveable.
Try a bicycle or a stroll instead of embracing the WALL-E.
If you feel that way about transit you may not have tried a good transit option like Hong Kong MTR with 90 second headways and travel from and to substantially everywhere you want to be.
>Try a bicycle or a stroll instead of embracing the WALL-E.
You see a robot driving around in a pile of trash.
I see a robot with nobody micromanaging him telling him how to live his life, etc, etc.
<we are not the same meme dot jpeg>
Well for my commute the trains are every 30 mins or so - pretty convenient times and a short walk from the office. The ticket is cheap, much cheaper than a days parking and during the trip I get to sit, look at the view and sip a coffee. The train is way more relaxing than the equivalent drive - which due to traffic levels at rush hour would probably take twice as long (at least) and be extremely unpredictable.
So when I have the option I'd rather take the train - of course I also drive a lot of places.
I think the answer to this is microbility bike/scooter sharing (ex: lime)
Trains to cover the longer distance and micro mobility options to get to exactly where you need to go
Fairly unreliably? Unlike cars, trains do not typically suffer from traffic jams.
This is based on my personal experience, I used to ride trains for travel a lot. I grew up in Europe and lived there for 31 years so this is not based on ignorance.
ok? Your personal experience is not the entire truth and never will be. Japanese trains are on time. Swiss trains are on time. That's not based on ignorance either.
I was in Zermatt last month and was unable to take the Gornegray Railway due to mechanical issues. Even Swiss trains have problems
You haven't been on the Washington DC metro, I take it. (Ok, you're technically correct, they're not typical.)
You haven't been in Washington DC traffic, I take it. It isn't like it is a choice between breezing to your destination unimpeded in your car to arrive reliably on time, and taking the variable, unreliable Metro.
I didn't comment on the road traffic at all, but good try.
Private cars seldom fail to work because the drivers are striking to reduce their hours to 32 hours a week like London last week.
Buddy the tube seldom fails for that reason either. Plus some self-driving sauce would reduce their hours to 0. Certain lines in London like the DLR are already driverless (Grade of Automation 3). Most of the other lines are GoA2.
Yeah, first time for that particular one but the transport not working for various reasons is not that unheard of.
I'd gladly take a Waymo from my home to the train station, zip around the country without traffic jams and hop in another robotaxi at the other end.
I live in Switzerland and commute mainly via public transport. We're very privileged here.
Because of decades to centuries of investment, holistic planning and expertise, we have one of the best networks in terms of quality, punctuality and density.
It's a plant the trees for future generations kind of deal, especially in Switzerland, because large, "flashy" projects are rare compared to to the more continuous and steady improvements, due to how funding and planning are set up.
Trains are one part of a larger transportation system. And they are very good at what they do. But you also need metros, trams, buses and so on.
And with that you can build a system where most places, including 50 people mountain villages are well connected.
I can go from Genf to a tiny village deep in the Eastern mountains with 4-6h. I can make that journey with no planning ahead what so ever.
Cars are actually restrictive. What if you want to have a drink? What if you are in a place that is different from your car? What if you are old or disabled? What if you are a tourist? What if you are not allowed to drive because of a traffic violation? And there are also these people called 'kids'. When I was 15 I went from Switzerland to Czechia with the train, no problem.
True freedom is to have a good public transit and potentially car as an option.
> which is not where you want to go
Oh the horror, you might have to walk a couple of minutes (probably less time wasted than circling around to find a parking spot, and then walking from it to your destination).
> at a time that is probably not exactly when you wanted to arrive
Yeah, no. Trains in properly developed networks are extremely frequent. At the off-peakest of off-peak (Sunday late evening), the RER near me is every 15 minutes. During peak hours it's every 5 mins.
I disagree that self-driving won't reduce traffic, at least from the perspective a Virginia resident. Commuting into D.C. is in theory very quick, except for when there are crashes. Crashes double the commute time, and there's _always_ a crash. This is pretty much the only source of traffic in my area. I think the primary benefit of self-driving would be lowering the crash rate, and as a side effect traffic.
This gets brought up a lot but I think it's missing some key points.
1) Being driven around is the best transportation mode for most of the US. It's very comfortable, private, fast, and point-to-point. It stops working well at very high density, but that level of density is only seen in a few places in the US. I'd like more people to live in dense areas but for the foreseeable future self-driving vehicles are going to be the best solution for most trips in the US.
2) At very high densities it's true that cars can move fewer people per hour per 10-foot lane than other modes and so you run into congestion. But that's measured with the current vehicle fleet and human drivers. With high autonomous vehicle penetration you could implement congestion pricing that encourages high throughput vehicle design. That means private vehicles that are much much smaller (think Isetta-like design) that can follow at very short distances. Along with the elimination of on-street parking we could see a many-fold increase in road throughput.
3) At even higher density levels the same congestion pricing mechanism would encourage people to use microbuses that would operate similarly to Uber Pool. Compared to today's busses they would have equal or greater throughput, be point-to-point or nearly point-to-point, dynamically routed, cheaper to operate and faster.
4) At the very highest density levels it's true that nothing can match the throughput of the subway. As others have mentioned, AVs are a great way to connect people to the subway. Many trips intersect with the highest density urban core for only a fraction of the journey. More people would take the subway if they knew they could get to and from the stations easily and quickly. AVs let you mix-and-match transport modes more easily.
Cities should start engaging with vehicle manufacturers to start getting these high density vehicle designs worked on and figure out the congestion pricing mechanism to properly incentive their rollout.
As with many "tech innovations" in the transportation space, this rapidly turns into reinventing the bus. #3 in particular is just "the bus, but more frequent" which you can do by simply increasing bus frequency dramatically, which most American cities should already be doing but don't, because of their budget priorities and the stigma of buses as something for poor people.
This rings less like some missing key points, and more like an entire, comprehensive traffic strategy. I'm not really sure what the point is meant to elaborate on. Maybe something like "Self driving cars in themselves wouldn't solve traffic, but well designed, purpose-built AV's combined with surge pricing and (when necessary, depending on the location and journey) trains/subways could do it." Did I understand you correctly?
My experiene in UK cities is that Taxis really come into their own at night when:
- The trains often aren't running (and there may not be the volume of passengers to justify running them)
- The road are empty so traffic isn't really an issue
As mostly a cyclist (I drive roughly 10% of my transport, the rest is biking and transit), my experience with self-driving cars is that I feel much safer riding in front of them. They're less likely to pass dangerously close to me to drive past me, they're less likely to tailgate me, they're also less likely to just drive me into the door zone, sidewalk, or a parked car. I'm a very confident cyclist but I suspect newer, more skittish cyclists would agree.
If you can restrict certain roads to autonomous cars (or heavily limit the number of non-autonomous cars) then you don't need to build as much bicycle infrastructure (a buffered lane is probably all you need, as opposed to bollards or true grade separation) and I can guarantee you more folks will feel comfortable riding bikes. This is aside from how frequently human-driven cars end up colliding with, damaging, or blocking non-grade-separated forms of transit.
> It's just boring because self-driving is much easier when you build the road to support it instead of removing all constraints and adding GPUs, lidar sensors, cameras and an army of fall-back operators in overseas call centers.
I do bike advocacy so this kind of rhetorical gotcha can make me feel good and hit the upvote button but in reality city councils and other elected officials are mostly people skeptical of the benefits of bicycling, worried that buses/trains would place too high a tax burden on their constituents, or deep down convinced in their lizard brain that Americans are too carpilled to ever do anything else. If you can change this by running for your local council, do it!
Don't get me wrong, we need more bike infrastructure and we needed it yesterday. But anything helps. I'd love to see certain corridors of SF be restricted to transit, autonomous vehicle, and cyclist usage only. Market is already only for transit and cyclists so there's precedent.
Bicycles are another way to address traffic, because they take up so little room and can be essentially free and often more convenient for shorter trips. Of course that means you have to have bicycle infrastructure where you don't have to run serious risks to your life every 3-5 minutes during your journey.
This statement is mostly wrong.
Cars as a shared service (shuttles, Uber, Waymo) absolutely solve traffic compared to personal vehicles. Shared cars have much higher utilization and require a lot less space.
I agree that trains are a fantastic way to move large groups of people, but a world with more shared cars (which may be brought about faster with Waymo) is a good thing for most cities.
Do I really care about traffic if I’m not the one driving in it? I guess if you’re looking at highly disproportionate delays but I really wouldn’t care about traffic otherwise.
> Do I really care about traffic if I’m not the one driving in it?
As someone who took the N across San Francisco every day for 5+ years: Yes, you would. Imagine a 5 mile journey taking 50 minutes. Even if you can nap or listen to a podcast, it's still a waste of time.
Trains still don't solve last mile transport for most people (even in places with robust transit systems)
Self driving cars might not solve traffic problems but they could greatly reduce them. Problems like traffic waves and gridlock go away when all cars are driving themselves.
The last mile is a solved problem. Most people can walk (and many of those who can't would need human assistance anyway). And then there are bikes, electric scooters, and other light vehicles that use space much more efficiently than a car.
Self-driving cars may help with the actual weakness of transit, which is the long tail of trips. Trips on routes with too few passengers to justify good transit service, and with the trips too long for the last-mile solutions.
Walking a mile with groceries, a baby, furniture, etc. is not really a solution.
I'm not saying self driving cars are the solution, but they are a piece of the solution.
Walking a mile with groceries or a baby is common. People in less car-oriented neighborhoods typically do quick visits to a grocery store when it's convenient for them several times a week, rather getting a week's haul of groceries in a single visit.
With furniture, you usually pay for delivery. Especially because the furniture store probably doesn't have the items you bought on site anyway.
When there are grocery stores within easy walking distance, people tend to grab the next few days’ worth of groceries. When everything is inconveniently far apart, people drive forever away to get giant stocks of things from Costco to haul back to their house.
If people aren’t spending $12k a year[1] to own a car, paying $50-150 to have a large piece of furniture delivered isn’t a big deal.
When there is well maintained, pedestrian friendly infrastructure, instead of a tiny uneven sidewalk inches away from 45 mph traffic, pushing your baby stroller home is not an issue.
[1] https://www.bts.gov/content/average-cost-owning-and-operatin...
Is it just me or don't people go on walks with their babies/children all the time? Also riding a mile with groceries & babies is trivial.
Cars are a piece of the transportation puzzle, but groceries and babies aren't why they're needed.
The last mile problem is only a problem because of poor layout. Build homes and work near transit nodes (instead of in the middle of nowhere) and there isn't a problem in the first place.
> Problems like traffic waves and gridlock go away when all cars are driving themselves.
How would that make those problems go away? It could probably slightly alleviate them in marginal cases, but any given road has a finite throughput limitation, and once it is reached, it wouldn't matter even if every robo-driver were perfectly synchronized.
> Trains still don't solve last mile transport for most people
This has not been my experience since moving to Manhattan last January. Subways, alone, close the gap between regional rail and most destinations astoundingly well. I haven't yet needed to use a bus (but they seem abundant, too), and I haven't even thought of taking a taxi yet.
Here, robust transit has solved the last mile problem for most people.
Sure, in areas without robust transit, transit is a problem. But I'm responding to RandallBrown's assertion that there's a persistent last mile issue in areas with robust transit. There's not. Manhattan is evidence that robust transit solves the last mile problem for most people.
Most cities don't have the density and wealth of Manhattan. How do we solve the last mile problem for everyone else?
This is the best map I could find:
https://cwhong.carto.com/viz/6dfca01c-47e5-11e6-9fd3-0ee66e2...
Weasel words are doing a lot of heavy lifting here. There is a persistent last mile issue even in NYC, even in Manhattan. You're right that in Manhattan most people can use the subway as a last mile solution. However that map hasn't changed much in quite a while. The subway deserts that exist (in Manhattan and the other boroughs) aren't going away anytime soon because building new subways is eyewateringly expensive.
The inflexibility means that even when the subway is a viable last mile solution it may not be the appropriate one. For instance I had to go from Ridgewood to JFK a few years back. I was maybe a five minute walk from the subway. But were I to take the subway from one end of Queens to the other I would've had to go all the way to Midtown and transfer to LIRR.
Hell I've generally had to rely on buses for last mile connectivity even in London which certainly doesn't suffer from a lack of subway service.
self driving cars will increase traffic as they remove barriers that prevent people that cannot drive from using cars, thus increasing the amount of cars on the road.
They replace taxis and potentially postal and trucking applications in future.
It’s certainly not a replacement for mass transit. US is sparsely populated compared to Europe and mass transit don’t work as well in the suburbia. That said, I do see many transit oriented development in SF Bay Area where high density buildings are being built near transit stations.
> Trains are self-driving. Europe already has the better self-driving system.
Well, I'm in Europe and it ain't here. Waymo can't get here fast enough.
> Trains are the only way to address traffic.
And how do you get to the train when it's too far to walk and you're not a cyclist?
Ideally: there's a train close enough to walk, or a bus or tram that's nearby that runs frequently, is clean, and doesn't get stuck in traffic because there's not much car traffic.
Slightly more realistic: enough people can and do cycle or walk to the train that pressure is relieved on the roads for those who cannot cycle or walk.
> And how do you get to the train when it's too far to walk and you're not a cyclist?
You get the bus, or you cycle, which is a life skill any able-bodied adult should have, not limited to cyclists. Of course not everyone is capable of cycling, but not everyone is capable of driving either.
A person living in DC can take the subway to Union station, take the Acela line to NYC, and then take the subway to their final destination.
The elephant in the room is rideshare commuting is for extremely rich people. Who else can afford the probably $75+ a day it costs on a two way commute?
Yes, driverless does not solve any real problem. When I come from work, I still have to sit in a car. Yes, I can work instead of drive, but that's only in theory because in practice the G-forces won't allow me to.
A robot cook, however, __would__ solve a practical problem for me.
Anyway, this whole approach is not even solving first-world problems (many families struggle to pay for a car), but it's solving the upper-1% of first-world problems, maybe. Except those people can afford to pay drivers who are now out of a job. So yes, what is this even solving??
Trains are not panacea some people here keep thinking they are. You would need to have train stops every few hundred meters changing it into some city subway or tram, interconnected with dense and fast local public transport.
I live in Switzerland, the place for trains, efficiency and its small and dense, an ideal situation right. Tons of people use trains every day, tons of people also bike for closer distances in good warmish weather but still highways are chock full and getting fuller every year. Public transport for out-of-city commuters is simply slower, often much slower.
This morning I was considering taking a motorbike to a train station that is 5km away, then 40 mins trains and 10 minute walk to work. I took the car instead for a change, I was faster despite having to cross the very center of bottlenecked and car-hostile big city (Geneva) in top rush hour. 65 mins door-to-door via public transport vs 45 in car. That's one way, meaning 40 minutes of my private life daily saved that I can spend ie with my kids and not staring in the phone or out of window.
Normally I take the motorbike if weather permits, if not I take the public bus to the train, adding additional 15 minutes each way. That sucks pretty badly. I doubt other countries have this figured out better, and not everybody can or wants to live in city centers, especially when raising small kids. We did it for 10 years, had a work commute of 5mins via escooter, but I rather have current commute and live and raise kids in small commune next to wild forest and vineyards than that.
All above is usually much worse in many parts of US.
Why do you hate buses?
[dead]
Mobileeye in Munich https://www.mobileye.com/blog/self-driving-robotaxi-sixt-ger...
Moia (Volkswagen) in Hamburg https://www.moia.io/en
Mercedes autonomous driving https://group.mercedes-benz.com/innovations/product-innovati...
Yes, but they said "meaningful".
There's some self driving tech being developed in Europe, but AFAIK nothing is at the current deployment level of Zoox or Aurora, let alone Waymo.
Does it matter where it's developed though? Once it's good enough to expand into all major US cities they could look into deploying in Europe too.
Im happy to let Americans be the beta testers
For the consumer, maybe not, other than a delay of some years.
In terms of having the industry? Absolutely. How many other areas of "tech" has Europe basically punted on and ceded to Americans? Currently there's some gnashing of teeth across the pond for how there's no real European equivalent to the big US cloud providers (AWS, Azure, GCP).
There doesn't have to be an equivalent of everything, I wouldn't want to use US cloud because of price and governance. At most I use the "cloudy" services and rent "capacity" from a European provider, companies are fleeing the cloud. They're done subsidizing Amazon deliveries.
MobilEye and Mercedes works on self-driving, so does BMW. It's probably not Waymo quality, but just because there aren't cars on the (wide and car friendly) roads doesn't mean nothing is happening.
Meanwhile Europe has solid infrastructure for electricity (esp France), ASML has no competition, Carl Zeizz is world leading in optics, there's probably a Leica LIDAR in the Waymo cars... I mean while we're throwing pies and bringing up other markets..
My old boss was working on a project with Leica where he was working with some partner on self-driving industrial machines, they we're using Leica gear for collosion avoidance and such.
Europe doesn't need self-driving cars, we have alternative modes of transportation. Where it's needed (mines and industry) it's already there. And whatever modern car you're driving here has ADAS which helps make driving comforable.
Sorry, but this is clearly just cope.
Yes, it's fine to give up the lead in any one subsector, but Europe is so far behind in tech industries in general. It's not just cloud services or self driving cars, look at SpaceX and Starlink: Europe has no equivalent to either, and is many years from gaining one (I'm aware of some plans, but they're far away from being able to actually launch, and some are dubious besides).
Both major smartphones OSes? Run by American companies. Major desktop OSes? Two by American companies, one originally started by a Finn, who still manages it...and he moved to Oregon.
But you don't have to take an American's word for it, just read Mario Draghi's report. The man loves Europe, deeply understands the European economy, and has a whole lot to say: https://en.m.wikipedia.org/wiki/Draghi_report
So any b2c thing where you're going to abuse your customers is American, what an achievement!
There's no denying America has done good in some industries, but when it comes at the cost of societies weak I can't help but think it doesn't matter.
SpaceX and Starklink aren't very important to me, I don't know who they're important for except Ukraine, boat and RV owners.
The report says we must invest in electricity infrastructure, well sure so the dude compares against China and USA at the same time? Crumbling infrastructure is the definition of USA 2025.
The cope is American Exceptionalism, we're doing just fine even though we're fighting a unprofitable proxy-war and missed all those b2c investments to leech off humanity.
There's no desktop OS from Finland, that's a kernel and yes he's now American as you guys usually were better at finding ways to turn good into profit.
> The cope is American Exceptionalism, we're doing just fine even though we're fighting a unprofitable proxy-war and missed all those b2c investments to leech off humanity.
We are not doing just fine. We have low economic growth. We are unable to beat off Russia's attack on Ukraine without American help. Germany has crumbling infrastructure just as much or more than America. We have not contributed seriously to any important innovation wave since before... 2000? The invention of the PC?
Everything isn't about economic growth, quality of life is more important for people.
EU aren't able to fend off Russia because other countries can't participate properly in the war.
EU contributes significantly to very many important innovations, though you seem to have made up your mind.
Economic growth is a measure of how much goods and services are available to everyone. If that isn't improving, that means your quality of life is lower, ceteris paribus. It means you don't produce enough energy on your own are dependent on Russian gas. It means you don't have enough surplus to sustain a military.
I assumed the parent was referring to "GDP growth" which doesn't matter when inflation eats it all and new coins go to megacorps rather than back into society, European standards of living has been consistently improving, especially for the poorer nations.
I can't defend Germany for refusing nuclear in favor of Russian gas, but at the time it seemed to some like a good idea to strengthen relationships through trade and encourage democratization.
It's a damn shame that we're buying Russian gas, it's hilarious that I keep hearing about this from Americans but not Ukrainians.
USA is huge. This is happening in a small part of the USA in a very limited fashion. It's not like the USA has driverless cars everywhere, 99.9% of the population never saw one.
I'd guess Waymo covers 5% now. San Francisco, San Jose, Los Angeles, Austin, and Phoenix are ~10% of US population. Waymo service areas don't cover all of those cities.
Considering tourism and people living just outside service areas who see them but don't get to use them (which includes me sadly) I would not be surprised if 10% of population had seen at least one.
> San Francisco, San Jose, Los Angeles, Austin, and Phoenix are ~10% of US population.
Surely you're describing metro areas? There's no way those five cities add up to 34 million people within city limits, given that none of them have 6 million people.
The MSAs added up to 27 million based on the 2020 census, so "close enough". https://en.wikipedia.org/wiki/Metropolitan_statistical_area
That said, Waymo's service areas are nowhere close to covering the full MSAs: https://support.google.com/waymo/answer/9059119?hl=en
- SF doesn't cover East Bay (two thirds of the MSA by population).
- Silicon Valley doesn't cover San Jose, and barely reaches into Sunnyvale (basically just covering the Google Moffett Park office buildings).
- The Phoenix area is missing most of the densest parts of Phoenix itself, as well as anything north / west of the city.
- Los Angeles doesn't even come close to covering the city, much less the rest of LA County or any of Orange County. (Maybe 2-3 million out of 13, from just eyeballing the region.)
On Uber (https://support.google.com/waymo/answer/16011725?hl=en) there's also Atlanta (which looks like it actually has very nice coverage, other than the western half of the city) and Austin (again focused on downtown / commercial districts) which help drive up the numbers.
The population that's had opportunity to see Waymo in the wild is probably higher because they're testing in quite a few cities now (a sibling commenter mentions NYC, for instance).
I saw one in New York the other day!
About 43% of the US population lives in 25 metro areas so Waymo doesn't have to be in a lot of places to have a big impact.
JYFI Cruise is “dead” after their SF accident 1-2 years ago. I believe GM has even written down their Cruise investment.
“More than 50 cities across China have introduced testing-friendly policies for autonomous vehicles.”[0]
Europe could do the same but they have other priorities.
0. https://restofworld.org/2025/robotaxi-waymo-apollo-go/
UK:
>pilots of self-driving taxi- and bus-like services will be brought forward by a year to spring 2026, attracting investment and making the UK one of the world leaders in this technology
I'm wondering how self-driving cars will solve the priority problem of narrow streets of UK towns where drivers need to let each other pass all the time.
I've wondered that myself. It seems quite challenging for human drivers at times. Around Ladbroke Grove you quite often get some complicated jam with two busses and about ten cars stuck.
Yeah I'll believe this when I see it. Most UK roads are significantly harder to drive on than anything in the US. That's why they always test these things in Milton Keynes.
Also a lot of UK driving requires communication with other drivers (letting people out, etc.) in a way that US roads don't. I'm not sure how driverless cars can handle that.
I really wish we could get them, because they're great. But I'd say we're talking 10 years behind the US simply because of the extra engineering challenge.
> making the UK one of the world leaders in this technology
Are they also planning on completely overhauling their economy and tax system to attract the engineers required to make this happen?
What do you think degrowth and decline means? Vibes and essays?
It's not just driverless cars either - delivery drones (e.g. in China), a lot of health tech (as they have more check-ups in the USA), Starlink, Neuralink, a space programme, etc.
I'll trade you your train network for our self driving cars!
> I’d really like to see either a Waymo competitor emerge in Europe, or even Waymo themselves operating here
I think you’ll see American and Chinese self-driving kit in Europe once it matures. It’s just easier to iterate at home, so while the technology advances that’s where it will be.
> As a European, I can’t help but feel a bit sad that we’re missing out on the driverless side of things
I don't know about other countries, but Spain will probably be one of the last ones to get it, thanks to the Uber-powerful (heh) taxi driver lobby
Maybe there just not enough interest? After all there is good public transportation (especially rail), increasing biking habits and just loving the driving experience.
you have buses and trains, you don't need waymo
I feel the opposite. Self driving cars seem like a meme because driving is fun and trains are better. If either of those premises is not true in your geolocation then self driving is not the solution either.
Cruise is basically winding down. Tesla is the other major competitor
Apollo Go (the Chinese Waymo owned by Baidu) is planning to start road testing in Germany and the UK in 2026, in partnership with Lyft.
Youre missing something very important. Train infrastructure in the US sucks. Not the case in the developed areas of Europe.
My personal use of a car has declined pretty dramatically the past few years. Trains are pretty good here in the UK.
I don’t think wages in europe are high enough to sustain this model business very well. When you track waymo deployment its in placed where plenty of high income price insensitive people are to be found.
Cruise is dead.
Wayve seems promising. I heard they want to open up in London soon
I don't remember any plans Waymo has announced for Europe, but they are testing in Japan.
Mercedes is quite close. They have demonstrated commercially viable Level 3 ADAS systems.
Regulation and under investment
Those darn regulators, don't they realise companies just want what's best for us?
There’s obviously a balance and EU’s arrogant regulators have far exceeded it
we have effective public transport in most major cities!
As an American with extensive time spent in Europe, I’d much, much rather have European-style metros and tramways than self-driving cars.
Waymo (though a technical marvel) is a bandaid over our inability to build and maintain public infrastructure. Be sure to cherish what you’ve got.
European cities have lots of taxis. Same with Asian cities. They will obviously have AVs in the future. I'm not sure why you think they should be mutually exclusive with transit.
Many American cities don't have the population density to make metros and trams economically viable. And those few cities that do have comparable density (New York, Chicago, namely) do have metros.
Public infrastructure has high overhead costs, and low population density means there isn't enough ridership to make it viable.
The problem is when cities treat car infrastructure as absolutely mandatory, and all other transport infrastructure (pedestrian, cycle, bus, tram, train) as optional. When you say that everyone has to be able to get everywhere by car all at the same time, you have to build more roads and parking (at minimum more roads using taxis, self-driving), more roads spread everything farther apart, which means more distance per trip, which means more cars on the road, which means more roads, which means everything is spread farther apart, rinse, repeat.
American cities low density is a direct result of designing for car infrastructure over all else. And car infrastructure is far more expensive than other transportation, and since increased car infrastructure lowers density, it directly makes all other transportation more expensive and less viable.
Since cars are the most dangerous form of transport, for other drives but more so for cycles and pedestrians, it makes it less feasible to use them for your first-last mile. Then you add in that, as the roads grow and distances multiply, speeds are increased to attempt to compensate, multiplying the danger to anyone not in a car.
Rotterdam — a city with a population of around 650,000 — has both a metro and a tram system. Extraordinary density is not a prerequisite.
And in any case, there's no reason that public transit needs to be self-funded. We don't expect the same of most of our other public services.
Rotterdam has 3,000 people per square kilometer. Contrast that with the San Francisco bay peninsula's ~1,100 people per square kilometer.
This is demonstrating my point about population density and transit.
The peninsula might not be dense, but San Francisco has a density of 7,194/km2 and the transit situation pales in comparison to Rotterdam's.
There are many urban areas in the US with population density of 3,000/km2 or higher that do not have any public transit at all.
People in the thread are asking why people ride Waymo to SFO, which is well outside San Francisco proper. Thus, the whole peninsula's density is what's relevant.
The US does not have many metro areas with population densities above 3,000/km2. And those that do, like Washington D.C, NYC, Boston, Chicago, do have metro systems.
American public transit construction costs are now ridiculous in terms of both money and political capital. Even somewhere as sprawled as San Jose now requires well over 1b/mi to build a subway under; BART could've acquired an entire autonomous driving company for the cost of the Silicon Valley extension.
As an American, I think you’re naive and short-sighted.
You must realize that, at some point, self-driving cars will be ubiquitous. Maybe not for 50 years, but they will be.
What you’re actually saying is “I’m virtue-signaling with Europe because that’s what the cool kids do”
…What? What sort of terminally online strawman would be spending his free time “virtue-signaling with Europe” to some anonymous bozos on a tech forum? What a dull and intellectually uncurious reply.
I think self-driving cars may eventually become common in areas where cars are currently common. I think public transit will continue to dominate in parts of the world where it currently dominates, because it is simply a superior user experience for the majority of people when the government cares to invest in it. (Not to mention far cheaper and more egalitarian.)
I am conveying my lived experience in most European cities I've been to.
> a superior user experience
A superior user experience is going exactly from where I am to where I want to be safely, quickly, and affordably. Self-driving cars are looking really good for those criteria.
$20+ per ride is affordable? Waiting 10m+ for your ride and slowly sifting through traffic is quick?
In London, Paris, or St. Petersburg, I pay a few bucks to hop on a train that runs every few minutes and rapidly end up across town, roughly in the area I need to be. It's literally the cheapest and fastest way to get from point A to point B, not to mention tested at scale and thoroughly battle-hardened over the course of a century.
Not every city has this privilege, of course, but surface trams are 80% of the way there, especially if they have right-of-way. And they don't make pedestrians' lives a living hell: https://www.youtube.com/watch?v=bNTg9EX7MLw
> $20+ per ride is affordable?
In the US, the unsubsidized price of a ticket is close to this amount.
> Waiting 10m+ for your ride and slowly sifting through traffic is quick?
In my city, it's difficult to pick any 2 points that are faster to get between by public transit vs. taxi.
Every city is different, but trains rarely make sense in the US (outside of NYC).
Right of way is the huge advantage of trains, it would be great if self-driving vehicles could have that same advantage.
Where did you get $20 figure? Self driving cars are bound to be much cheaper because there is no human needed to drive it.
you're shouting at a wall here. This site is absolutely infested with US techbros who believe that the solution to any problem is zero regulation and more computers.
This is why self-driving cars appeal to this crowd. You and i seem to be from a world where public infrastructure like clean, affordable transit is the goal. This raises the floor for everyone. Many here would rather think solely of their own comfort, which is fine, but despite repeatedly being told that they are short-sighted, they refuse to change.
The core for a good experience is a good structure.
In many regions of the U.S. people live too far apart, shops and businesses are zoned apart into wide spread business areas. Public transport won't provide a good experience.
In a notable part of European cities people live in denser quarters, where a "third place" is reachable in walking distance, some degree of shipping, doctor visits, work are close by. There public transport can fill the gaps for the remaining trips in an (space) efficient way. Self driving cars however would clog the area.
Adapting US settlement structure to allow public transport won't happen. However a self-driving car can turn the dial for individuals to move out of the urban European area into more rural areas. Question is how big that group is.
Try moving a few bags of sod and mulch via public transit. Condescending tone is condescending.
Designing our urban transit around the needs of the mulch-bearing 0.1% seems like a bad idea.
0.1%? You think so? Sorry you're wrong. Suburban population is the largest demographic in North America.
And Home Depot says otherwise. They have reported record profits year over year for the past two decades. Just because you don't use sod in your condo doesn't mean suburbanites don't need it for their homes.
Throwing out bullshit statistics like 0.1% is an ignorant take.
See, it’s super easy to be a jerk.
Saying we shouldn't design around that use case isn't being a jerk.
And the exact number wasn't the point. The percent of consumer vehicles on the road that are carrying a significant payload to/from home is pretty small. Especially areas where transit even halfway makes sense. What's your best estimate?
Where I live, the percent of cars that carry a load that would be unwieldy to manage on public transit at least once per month has to be at least 50% and probably closer to 90%.
From Costco trips to babies to wagons, strollers, wheel chairs, hardware stores, bigger box purchases like a TV, out of town trips to visit friends, pet grooming, airport trips with luggage, it's hard for me to imagine a life without a car.
I know you can just say that I'm a product of my circumstances and culture and you don't need a car for any of that, or there are other ways to accomplish my goals, but I could say the same back to you. And the arrow of time seems to point to people everywhere moving in the direction of wanting personal mobility whether horses, bikes, or cars.
It's not all or nothing, but it seems to make sense to me to build around cars as a first class concern, in addition to other forms of transit. Some places in Europe obviously can't, for historical reasons, but I don't see that as a benefit per se, so much as something to have to work around.
Edit: I should add, I did live car free in Boston for 10 years and loved it and didn't really perceive any shortcomings at the time, and even hated having to buy a car when I moved. But now in my 40s with two young kids and a house and an elderly mother, it's an entirely different situation and I can't see how it would work. I would suggest if you're totally anti-car but only in your 20s or early 30s, your opinion might change as your circumstances do. I also lived for a year without a car in Singapore and that was tolerable in a way that wouldn't have been in most places, since it has some of the best public transit in the world, but even there cars are considered luxuries and it would have made things a little easier.
No no no, not the percentage of cars that sometimes carry a load, the percentage of cars on the road that are currently carrying such a load.
If you do that once a week, then you can use transit the other 90% of the time. If people use transit 90% of the time, then we can build smaller roads and de-prioritize cars. That's the argument here, that transit can dominate in co-existence with self-driving cars, not that we'd need to get rid of cars. And especially in the context of waymo there's no effect of "I'm already paying a ton of money to own and insure a car, I might as well use it every trip".
(And again, this is in moderately dense areas where transit works and you actually care about how many cars are on the road to begin with. And it doesn't have to be 90% in particular.)
EU’s amazing infrastructure is the Minitel that will prevent it from getting the internet of self-driving.
Subways don’t solve last-mile problems or trucking.
>Subways don’t solve last-mile problems
We don't have a last mile problem, we have legs for that.
Good. Cars ruin walkable cities, and the last-mile problem can be solved in other ways.
And it's not just the EU. I'm sure that e.g. China and Japan will continue to invest in their excellent public transit infrastructure even when there are more self-driving cars on the road.
Much of Japan's transit infrastructure is private. There's nothing special about transit that means the government has to own it; being a government, it can regulate things without owning them.
Americans have this idea that transit is for poor people, which translates to "it's not important for transit to make money", which translates to "we need to make it illegal for transit to possibly try to make money", so there aren't even vending machines at the platforms. Whereas in Asia they do profitable land development at the transit stations.
> Much of Japan's transit infrastructure is private. There's nothing special about transit that means the government has to own it; being a government, it can regulate things without owning them.
Japan's private transit infrastructure is only private in high-very high density environments (inner-city) and subsidized in low-density environments (rural, cross-country). Ultimately private group transit requires population density above a certain threshold to be viable.
Cruise has been out of business for almost a year I think.
In Switzerland the Airport has 28 trains per hour that connect it directly to almost every part of the country. In addition to that there is a tram line and many bus lanes.
But I guess in SF they can take a taxi that might be a little cheaper because the company operating it is fine with losing 100s of millions a year.
Waymos aren’t even really cheaper than uber lyft and traditional taxi.
Elections have consequences. Your lawmakers won’t even let us use browser cookies without permission.
I think navigating European roads is a massive step more difficult than US cities. They've got wide lanes and a really strict grid layout generally. At least in the European cities I'm familiar with we have much narrower lanes, residential areas with parking turning 2 lanes into 1, old towns, and layouts that are completely unpredictable. Maybe I'm wrong but I think this is the bigger hurdle than regulation.
US and China basically.
Wayve?
One thing you are missing out on: mandatory loud (97 to 112 db) 1000 Hz audible beep when the vehicle reversing, oh so slowly, such as at the recharging station. Also, constant shop vac five horsepower vacuum cleaner sound. BEEP BEEP BEEP BEEP BEEP. VROOOOOOOOOOOOOOOOOOOOOOOOOOOOM
Oh wait, you thought those would be in the middle of nowhere? Nope.
https://www.karmactive.com/waymo-charging-noise-blasts-112-d...
It is not mandatory that backup alarms be 97 to 112 db. They only need to be "above the surrounding noise level". The loud beeping alarms were installed on most vehicles because most of them operated at loud constructions sites, so needed to be louder than that. it was easier to just buy the loud model to CYA, even if it was a delivery truck. They also don't need to 1000 Hz or to actually beep. White noise backup alarms are allowed, and in use in many delivery trucks now, and make a sound attenuated above 4000 Hz, which is much more localized and dissipates over much shorter distances. Waymo could absolutely have installed 85dB white noise alarms but chose to install 112dB beeping alarms. This is not a regulation problem.
Unless and until those noises that you mention are as annoying as those made by present time ICE vehicles, your point will remain irrelevant.
From the link:
These backup warning systems operate at approximately 1,000 Hz, producing sound levels between 97 and 112 decibels.
Santa Monica’s municipal code adds another layer of complexity, prescribing exterior noise limits of approximately 50 decibels during the day and 40 decibels at night.
The continuous operation—with vehicles reversing dozens of times hourly, including during late-night hours—continues to challenge community peace.
So, constant car screaming BACKUP BACKUP BACKUP I'M BACKING UP HERE right outside your residential window. Kinda sucks. A whole lot.
Ironically the car is virtually purpose built to not run over people, and likewise has an extensive sensor suite to detect people around it.
I suppose regulations don't care if you can see no one is behind you.
> It seems like most of the meaningful deployments are happening in the US
Because they are.
Across Europe you can randomly encounter a major town with a taxi cartel still blocking rideshares, as if its 2012
Don't worry, we're missing out on a lot of "progress" on this side of the ocean thanks to Trump's dislike of wind farms and RFK Jr's whole anti-vaxxer thing
We can’t even use Waymo when we land at SFO for a visit
Isn't that what this article is about?
As a fellow European I'm quite happy that these driverless POS's are not here. I can't even understand how int the hell are they legal over there.
The setup at SFO is currently quite annoying (Lyft/Uber require you to walk 5 mins to the garage roof, and drivers need to park/wait 5-10 mins away, so there's always a substantial delay). Taxis get the privileged parking spot immediately outside arrivals, but if it's busy you might still need to wait a bit.
I've been wondering for a while why Waymo can't offer a semi-managed solution to SFO to dynamically manage load, have just the right volume of cars inbound, maximize parking utilization, etc. with all of the nice intelligence that an app-based system would enable.
It feels like you should be able to have a buffer of cars waiting right at the curbside, and automatically refill that buffer on short notice depending on observed or predicted demand.
As an Uber rider, I actually love the SFO setup. The walk is short enough, there's actually enough space even during most busy times that there's no crazy honking of drivers trying to get in or out of the pickup zone.
Compare that to the mess that is Uber pickups at JFK, where you have big delays _and_ very poor traffic controls in and out of the pickup zones.
LGA does it way better, but the walk is a bit longer.
>I've been wondering for a while why Waymo can't offer a semi-managed solution to SFO to dynamically manage load, have just the right volume of cars inbound, maximize parking utilization, etc. with all of the nice intelligence that an app-based system would enable.
Uber could in theory do all those same things too, right?
To some extent, but I think it’s easier to have fully automated buffering if you physically control the cars. Eg you can have backup vehicles parked indefinitely nearby if you want, whereas there will always be some unpredictable churn from human drivers eg unexpectedly clocking off.
No idea if these are first-order effects in practice.
Taxis have a powerful local lobby; Google/Waymo doesn't.
I’ve never had to wait more than 5 minutes at SFO I don’t think and the system seems ok to me
counter point, I love the taxi setup, I wander out, no pre-planning, walk across the street with my headpones on and get in a car, my company pays for it. I suually pay more on uber or lyft, and it's faster and I don't do anything but walk from the plane to the car
I did this in Eastern Europe one time, ended up being made to pay 60€ for a ride that's 10€ on a ridesharing app (even with the "licensed taxi" option...) When there's reasonable price controls it is convenient though
I'm surprised and incredibly impressed at this announcement. It seems trivial, but the general feeling in the industry has been that SF would fight tooth and nail against robotaxis at SFO.
Probably because SFO felt the heat after Waymo acquired SJC approval quickly: https://sfstandard.com/2025/09/05/phoenix-has-airport-robota...
Most likely both agreements had been in negotiations for a while and not something they just pulled together last week in response to SJC, although it's possibly they could have used it as leverage (hey we've talking to SJC ...)
the major difference being that SJC is easily accessible by surface streets whereas SFO mostly isn't.
And Waymo doesn't currently operate on highways for passenger service (I think they have authorization to, but they're only testing on highways right now).
They should be able to get to SFO from Millbrae Ave and San Bruno Ave without getting on the highway proper, although it'll likely be a lot slower unless you're getting a ride from nearby. While SJC can serve downtown SJ and Santa Clara without getting on a highway.
Yes, but SFO (and the city of SF) has a long history of being hostile towards Waymo: https://missionlocal.org/2023/11/waymo-rebuffed-by-sfo-sf-gu...
What likely happened now is that SFO got a kick up their backside from the Mayor after the press started asking why it was still dragging its feet, while SJC approved Waymo swiftly.
I genuinely think things have changed with Lurie as mayor and 6 growsf endorsed people on the board.
It's going to take a long time for SF to overcome the reputation it built for itself in the 2010s.
Recent changes in the composition of the San Francisco Board of Supervisors (i.e. Peskin being out of government) may have something to do with it being easier than expected.
The NIMBY/landlord supervisors who controlled SF, such as Aaron Peskin and Dean Preston, are now a thing of the past.
Peskin is now reduced to showing up at protests with signs saying the rent is too low.
https://x.com/agarwal/status/1966365908085125384
Waymo got approval for SJC last week. That probably accelerated approval for SFO, which had been stalling. Nice.
When they get clearance to drop people off at the main terminals, that will be more convenient. Pickup at the terminals is harder. There will be a need for a staging area somewhere in the parking structures.
Few major airports I've been to allow Uber/Lyft anywhere near the pickup area, so many fliers are already accustomed to walking a quarter mile or so to their rideshare. But their inability to use the drop-off area is a new inconvenience, and I can see it limiting the appeal.
Waymo will probably get access to the drop-off area after a while. One step at a time seems to be the Waymo way.
Waymo at airports could work really well with automatic dispatching. They already have an app running in the customer's phone. It should be aware of when someone with a reservation gets off an airplane, and how close they're getting to the pickup point. With good coordination, as the customer heads to the arrival lanes, a Waymo pulls out of short-term parking and heads for the meeting point.
A few more years, and humanoid robots will put the luggage in the trunk.
this can be done today with humans. this is just dumb
Humans need to be paid, and often demand tips.
You think the robot is going to be cheaper than the human, leaving money on the table? They aren’t even cheaper than taxis or human rideshare.
> Few major airports I've been to allow Uber/Lyft anywhere near the pickup area
Few major airports have Waymo at all. Phoenix has allowed pick-up at the airport for ages. (EDIT: Never mind.)
I'm talking about Uber/Lyft drivers being required by many airports to pick up away from the normal pick-up area, usually down the road a bit or in a parking garage.
Harry Reid International Airport (LAS) is in a parking lot.
What airports are you flying out of? Every major airport i have been to in the last year has a dedicated rideshare pickup lane.
Uber black and at least lyft extra room have no problem picking you up at the arrivals
It’s wild that $goog is so undervalued (p/e 27) given Alphabet owns Waymo in addition to everything else, and yet Tesla is so overvalued (p/e 243!!!) despite zero Robotaxis in the near (or far) future and lackluster sales.
Goes to show empty promises and fraudulent showmanship sell better than actual working products that people use.
GOOGL is up like 25% over the last few weeks after they resolved the DoJ lawsuit about Search bundling. Clearly there were some investors who thought that was a material risk to the business.
Tesla is clearly a meme stock though, and an example of how the market can say irrational longer than you can stay solvent.
I finally capitulated and bought a few shares of TSLA, shorting wasn't working.
Err... that sounds like gambling sir
That's why passive index funds and a hands off approach are so often recommended. You cannot really mess up much by buying the whole market and then sitting on it long enough.
Unless you buy the S&P 500 and like half of it is the Magnificent 7, partly driven to that proportion by meme stocks and options gambling.
Wait until you read about options!
Yeah "that'll go up" but now you have to know "when" and "when do I stop that bet".
So you think it's going to raise further?
No, they were just scared of missing out.
I stuck some money in Uber at 90 recently based on fundamentals. I might be wrong but at least I used a calculator to be wrong :) and will learn.
It needs to beat Sp500 to be considered right.
Saying it’s “right” based on outcome alone is like saying ~half the people in Vegas and betting on black made a good decision. You can win and still have made a poor decision.
A better approach is to look at the full range of your bets and try and decide if the betting strategy was good. But that gets difficult when you consider outcomes are linked through wider economic trends.
If you buy Alphabet stock you're betting on the whole company doing well.
Google makes around $300B a year. Uber's entire business makes around $50B and that took a decade. Waymo would have to become a major business to move Alphabet's stock price in the near term.
Considering Waymo is very likely losing money, experiment very slowly with scaling up, and still raising billions in private capital outside Google... idk. Doesn't seem as simple as buy $goog in 2025.
Otherwise I agree Tesla is a bit of a meme stock.
I think Waymo has huge potential for being much larger than Uber - people are willing to pay more compared to ordinary uber drive just to avoid dealing with taxi drivers and tech will only get cheaper.
More than that, I think the ride-hailing business is just the fist volley in the self driving vehicle space. It’s a short jump from there to self driving trucks, self driving package delivery, self driving private vehicles, and on and on.
All of those spaces are actively being explored by various companies.
Can any of those companies catch up on self-driving faster than Waymo can pivot to their niche? Cruise seemed to be a distant second, but did themselves in with an attempted cover-up.
There are already self-driving trucks on the roads. Their pilots came earlier, because the problem space is much smaller.
They don't need to "catch up" to Waymo, because of the niche.
https://bigrigs.com.au/2024/04/18/driverless-trucks-trial-be...
> There are already self-driving trucks on the roads.
2 trucks?! I suppose that's the minimum number required to make your pluralization correct.
I will stand on my earlier statement regarding this particular outfit: they'll need to catch up because Waymo started class 8 variants in 2021 https://waymo.com/blog/search/?t=Waymo%20Via
That article also mentioned previous trials from other companies that are ongoing, from previous years.
And Volvo rolled a class 8 as well.
I see Australia in the article and pardon my rampant scepticism, simply don't believe it.
Lo and behold:
>A six-month trial of driverless trucks on public Victorian roads has been put on hold just hours before it was meant to begin after the transport union labelled it “shambolic” and “sneaky”
> "the futures of our truck drivers are jeopardised due to this poorly executed plan."
> “It’s unacceptable that these trials are being pushed by corporations that continue to disadvantage our hard-working mums and dads that work day in, day out to carry Victorians.”
Now this sounds far more like the Australia I know.
Looks like the entire trial was scrapped due to union pressure and never resumed. Same reason we can't even have Driver-Only Operation on NSW trains, despite specifically purchasing DOO trains that operate safely worldwide.
https://ia.acs.org.au/article/2024/-shambolic---victorian-dr...
And plenty have failed. Perhaps a smaller problem space but still really, really hard. Some self driving freight company failures: Starsky, TuSimple, Embark, Ghost, among others.
One promising self driving truck startup, Aurora, was forced to put a safety driver back in the driver's seat after testing in May.
https://www.ttnews.com/articles/aurora-driver-back-in-seat
"Forced" by the truck maker, who was forced by their insurance company. All these companies will face that hurtle. I suggested to my girlfriend, who is a corporate defense attorney, that she get involved in this area of legal practice. It's a legal minefield.
It does seem very messy! Will be some interesting precedents set over the next few years I imagine.
Probably not.
Cruise was nixed by GM execs, whom I believe were looking for whatever excuse they could find to shut the operation down. They simply couldn't afford to stay in the game for the long haul. Cruise was under pressure to appear more capable than they were, and they took risks.
Waymo is distinguished in that it doesn't need to pander to nervous investors to keep getting money. The company is Sergei and Larry's baby. Google's founders will ensure that Waymo is patronized until it can stand on it's own.
> ...I believe were looking for whatever excuse they could find to shut the operation down
Cruise's self driving license was suspended because humans displayed poor judgement by omitting from the official report details of their stopped car dragging a knocked-down accident victim under the car for dozens of feet. They took "risks" alright, and their harebrained cover-up was discovered by chance by the oversight body.
I believe any driver who covers up the details of injuries in an accident permanently lose their license, because they'll definitely do it again. What good is a self-driving subsidiary that can't operate on public roads?
I agree which is why I love that this is technically bait about various techs that want to claim/market to be 'Full Self Driving'.
Buy a Comma.ai and install it in a supported vehicle, and just try it out. It doesn't talk to GPS, but it handles left right gas brake on the freeway well enough, and that's with two fairly shit optical cameras and a radar system. Granted, geohot helped start the company, and he's no slouch, but if their system is that good, a couple things are true. A) Lidar isn't necessary b) Extensive mapping that Waymo does also isn't necessary c) that last 10% gonna take 500% of the time to get to L3/4/5 autonomous, and that last 1% is maybe never. The other day I was in a Waymo, and there was a semi totally blocking the street, backing into a loading dock. The Waymo correctly identified that there was an object in the way, and stopped and did not plow into it. At first it crept up to the semi, blocking it from making progress as well. It might have started backing up, I've seen them do that, but I was already on the customer support line as soon as I saw the semi blocking the road.
Comma.ai is probably the purchase I'm most happy with this year (to be fair though, I buy a lot of crap off Temu). Drives are now just "get on the freeway, and just chill." Pay enough attention because it's not collected to GPS and just in case something goes wrong. So to be clear, Comma.ai is not autonomous driving, it's classified as an ADAS, advanced driving assistance program. It just makes driving suck that much less, especially in stop and go traffic, for $1,000, and compatible with recent vehicles that have built-in lane guidance features. Waymo's got to be light years ahead of them, given how much money they've spent, so it's my belief that Waymo's taking it very slow and cautious, and that their technology is much more advanced than we've been told.
How does self-driving package delivery work? Who delivers the package?
There are several “last meters” delivery robots developed.
Short range drones are being used in Australia.
And I heard of at least one company working with apartment architects to standardize a “port” on the building exterior to which a truck/robot would connect to “inject” packages to the inside.
> "Short range [delivery] drones are being used in Australia."
Last I read (late 2023 IIRC) these were being cancelled in various areas, if not everywhere? People in neighborhoods were getting annoyed by the noise of drones buzzing overhead.
Like some sort of "mail chute"?
This was just an acquaintance some years ago in SF, but I recall it was fancier with conveyor belts and a protocol for the robot to communicate the size and weights of the packages being delivered.
Tiny catapults. It's the only correct answer.
Sadly, this would still be an improvement on many smaller delivery services that especially Amazon is fond of using.
The slaves obviously.
But to be serious, there may be a way of doing it, it just seems very far off unless you're talking about Amazon hub or something like that, where it would be more feasible (but still difficult to achieve).
Think of Waymo Driver as the equivalent of Android for vehicles. It's an operating system and a suite of cloud services for both autonomy and ride hailing.
The long history of "First mover advantage" being a myth implies they are more likely Nokia or Blackberry than Android
What about all the expensive hardware, gpus, lidars? That’s like having iOS on your phone and if you want android you need to buy extra things that are worth same price as your phone.
It might be like that if a Tesla Robo taxi could actually operate like a Waymo
And costs should be lower in the long run if you don't have to share the ride fee with a driver (not case yet because seems like they still have alot of staff to manage the cars)
Statistically Waymos are more expensive than Uber rides, but practically as an individual they are often cheaper than Uber, its very easy for the stated price to be lower
So its not even about willingness to pay more
Gig drivers are cooked
A lot of times the Waymo is only a few bucks more, so if you were going to tip the uber driver it balances out anyway.
I'd still choose Waymo if it was 100% more than Uber, the experience is so much better and I feel so much safer.
You might, but most people wouldn't, and more to the point, overwhelming more people will choose to drive their own car (or take transit) vs either Uber or Waymo.
If Waymo can drop its price by 50%, it could steal a lot of demand from normal cars and transit, but that doesn't seem like it's even on the conversation right now.
And the Waymo can't be a creep or sexually harass you.
PS nice name.
Driving in a car that doesn’t smell like driver just farted right before picking you up is worth the premium.
Ehhhhh maybe in some spaces...
I would need to see Waymo be able to handle something like Southeast Michigan before I could even get comfortable with trusting it to get me ubered t/o from home for maintaining the vehicle I need to commute when I can take a remote day or two...
And then also delivering that for a good cost.
I put it that way because, I do tip Uber drivers well (unless they cray cray) and they would need to properly 'undercut' uber with whatever model they serve up in more complex areas.
I’ve seen people claim, I need to say Waymo working In NYC, Chicago and other places but never Southeast Michigan. What’s so unique about that area?
Waymo works in SF Chinatown btw, which is probably the most complicated locality in its driving zone.
Why is southeast Michigan difficult to drive in? I don't know anything about the area but I would guess if GPS navigation works and it's less dense than SF/LA, most of the major issues are solved?
Waymo doesn’t own manufacturing of vehicles.
*for now
Who are these people?
There is no downside to having someone drive you Uber has homogenised the experience.
Anyone who's taken enough Ubers and/or has had bad enough luck to have gotten a terrible Uber driver. Pretty much everyone I know, along with myself have had multiple awful Uber driver experiences.
Did uber/lyft get radically better in the last 12 months?
I had one rapidly cycle their prius between 50 and 70 on the freeway because regenerative brakes save gas (I felt carsick for hours after arriving at my destination), and another actually get an angry mob to tap on the windows and berate their driving. (The mob was justified.)
Since then, I’ve given up on using them whenever possible.
> rapidly cycle their prius between 50 and 70 on the freeway because regenerative brakes save gas (I felt carsick for hours after arriving at my destination)
Weird take to me, unless you were on a lot of hills; at least in my Maverick [0] 55-65 is 'ideal' MPG range for long trips, going between speeds tends to trip things up and actually -avoid- the weird 'battery has enough juice where we just kinda lug the engine' mode.
Doing regenerative 'braking' compared to using physical brakes, absolutely can give energy for momentum/acceleration and save on the physical brakes wear and tear, OTOH any normal cyclist would say it's better to 'maintain' a given output power vs allowing deceleration and then going back up to speed.
As for why, well I'm not a physics person, but in general it's that you are having to overcome the rotational mass/etc of the wheels (i.e. tires, axles, etc), and no regenerative braking within the current laws of physics will make slowing down and speeding back up more efficient, at least on a flat road.
[0] - OK It ain't quite a prius but it works fairly close aside from overall drag...
Checkout this thread for who those people are: https://news.ycombinator.com/item?id=44258139
That TC article doesn't substantiate its overly broad claim. "People" aren't paying more, in general, across its US markets; it only shows that a subset of its customers in what is already the top-5 most expensive cities (SF) in the world are prepared, and at that, only 10-27% are prepared to pay significantly more ($5-10). Still fewer than the 40% who would pay “the same or less.”
Quoting: "Perhaps even more striking is how people answered a question about whether they would be willing to pay more for a Waymo. Nearly 40% said they’d pay “the same or less.” But 16.3% said they’d pay less than $5 more per ride. Another 10.1% said they’d pay up to $5 more per ride. And 16.3% said they’d pay up to $10 more per ride."
There are going to be lots of causal factors: number of rider(s), time of day, safety, gender, wait time, price estimate, predictable arrival. Let's see an apples-to-apples comparison/regression breaking out each.
I think waymo actually has a better km/accident ratio than the average driver. Plus if you haven't done it before, it'll be a cool experience to ride in a car with no driver!
But in the long term I think the point of waymo is that it'll be cheaper: no need to pay the driver if there isn't one!
Women. Turns out, Uber/Lyft can't really do anything about drivers assaulting passengers.
The words women and woman appear exactly once each on this thread. If there's one thing tech product management needs, it is to ask a woman. This is the most obvious blind spot in tech.
Maybe with the HN readership, but in general the public don’t want to drive in driverless vehicles and don’t want them on the streets. It’s going to be a long uncertain road for them to be accepted.
https://newsroom.aaa.com/2025/02/aaa-fear-in-self-driving-ve...
I don’t think Waymo is very likely a losing money experiment. I give them a 50% chance to be successful within the next 10 years. Successful being that self-driving cars are able to operate in 50% of the world/terrain types/region types, probably within another 10 years to scale up.
To all: also think of the productivity boosts. Working in the car or just napping in the car.
In the Netherlands this is already sometimes possible if your work is close to a train station while your house is too and you don’t need to switch trains. It’s a boon to be honest.
My favorite is the train from Amsterdam to Berlin.
Of course, if you carpool then you can do this too. One time I rode in a car as a passenger from Berlin to Prague while working the whole time. When we were there, we went to a DnB festival and we got back on the weekend.
Hybrid working is awesome :)
And self-driving cars could make it more awesome
They have already spent an enormous amount of money. It’s hard to see how they could make it back quickly, if ever. I’d like to be wrong, but I expect they will continue to be a money losing experiment for a long time yet.
How much money they've spent in the past is irrelevant. That money all came from investors, in exchange for a stake in the company. It never needs to be "paid back". Besides which, those investors have earned all those funds back already, and then some (on paper).
All that matters at this point is how much money they'll lose/earn in the future. There are no shortage of investors willing to put money into this effort, and they're growing exponentially, so there won't be any pressure for them to turn overall profitable for several more years.
Boeing may never make back the development costs of the 787. That was an absolutely epic disaster of a project. But that doesn't mean Boeing shouldn't build and sell every 787 they can profitably sell.
If Waymo is at breakeven including capex, opex, and overhead, operations logistics becomes the limiting factor. While Alphabet is capable of investing more money into Waymo, I think they've reached the tipping point. If you see Waymo expansion accelerate, bet on that tipping point having been reached.
How much money do they make off the average person in the value of ads shown per year?
Now compare to how much money the average person spends on driving per year.
If Waymo winds up running half the market in autonomous transportation over the next several decades, it'll make search look like peanuts in comparison.
You need to consider profit margins. The cost of showing somebody an ad is very near $0, which is what makes digital products so profitable. But when you do things in the real world, especially in highly competitive markets where the customer is extremely price sensitive, your profit per mile is going to approach $0. For instance WalMart's profit per item sold is less than 3%, and for driving this will likely be substantially lower (given the combination of customer price sensitivity + competition). The way you make up for this is in massive volume, but Waymo for now remains a heavily ringfenced operation and so it's not entirely clear how they reach scale. Google also has a very poor record of long-term performance in competitive markets.
The winner in self driving will likely be enabled by extreme vertical integration - you want to be building your own cars, cleaning your own cars, repairing your own cars, and so on.
The average American sees something like $500 of ads that go through Google per year. There's a profit margin of around 50% since Google has to pay publishers and pay for running search. So that's $250/person in profit per year.
The average American spends something like $12,500 in car+taxi/rideshare per year. Suppose with Waymo that goes down to $7,000 and it's 20% profit. That's $1,400/person in profit per year.
Obviously it gets much more complicated -- the profit margin depends on whether there are serious competitors to Waymo and how much Waymo's head start matters. Waymo will bring costs down further with shared vans and buses on demand. Profitability will rise with video ads in vehicles that you pay not to see. And so forth.
But autonomous rideshare is going to be larger than search any way you look at it. Profits won't be as high as search, but the barriers to entry are so high that profits will be high for a long time.
Those data you referenced are per household, not per person, and the majority of that is loan+insurance. The actual cost in terms of maintenance, fuel, etc is quite low, and that's the price that eventually will be the goal line for autotaxi companies. 20% [net] profit margins do not generally exist in competitive real world industries, outside of perhaps something like real estate. A net profit margin of 5% would be huge, and I think it will likely be much closer to 1%, or even less, simply because in the end it's going to be a commodity where all that matters is price.
I also think you're overestimating the impact of things like ads, buses, etc. The second Waymos become less pleasant than any remotely comparably priced option, they will lose customers.
No it's per adult not per household. The average household has 2.2 cars, so the figure per household is much higher. And it doesn't matter what proportion is loan vs insurance vs maintenance vs fuel, because Waymo replaces literally all of it.
And yes I assume Waymo will have high profit margins for an extended period of time because they have such a massive head start, and for a long time will be competing primarily against rideshare with human drivers, so won't be pushed below that. Their marginal costs will be much cheaper than that, not having to pay drivers. Hence 20% is not unreasonable.
Then, even in the long term, the economies of scale they develop and network effects will continue to give them a significant advantage. Not 20% margins, but way more than 1%. Especially as they start to vertically integrate the hardware at some point.
Here is where you would generally cite sources. [1] Those are the data from the BLS. Total transport spending per household is $13,174. The term they use is consumer unit, which you may have conflated with consumer/person, but it's practically the same as household. There are 134m consumer units, and 131m households.
Waymo is currently charging substantially more than Lyft/Uber and is not profitable. Human drivers can taxi in anything with 4 wheels and a hood, and its 100% their responsibility to take care of their vehicle, fuel it, clean it, and so on. Each Waymo currently costs ~$200,000 and is going to have a proportionally higher maintenance costs, and all of those costs must be covered by Google. So their costs are far higher than you're ballparking.
As for competition - Tesla has already launched a live robotaxi trial in Austin, so it's already here.
[1] - https://www.bls.gov/news.release/pdf/cesan.pdf
I was just going off the top Google result based on AAA data. Took a closer look and it turns out it's the average for new cars [1], so the discrepancy must be that your statistic takes into account the secondhand market. Thanks for the correction.
In any case, the overall point is the same -- it's a vastly larger market than Search. And what Waymo currently charges, and the current cost of their cars, is irrelevant. Waymo's business model isn't based on the economics this year or next year. It's based on the economics ten and twenty years from now, when costs have fallen dramatically as they switch to cheaper models and gain massive economies of scale.
As for Tesla, it's hard to take seriously given all the promises it's made and completely failed to deliver on. Their trial currently has a safety human in a front seat and is limited to a tiny group of testers. It's so many years behind Waymo already, and it's unclear if the technological approach it's taking will ever be able to catch up or meet minimal safety requirements.
[1] https://www.nerdwallet.com/article/loans/auto-loans/total-co...
Can’t imagine Tesla will be able to remove the passenger seat safety monitor any time in the next 5 years. Refusal to install lidar means Tesla’s AI has to be 100% perfect, which won’t happen for a long time, if ever.
The entire global taxi market is ~$250B a year.
Google made ~$265B from its ads last year.
Not the global taxi market.
The global driving market.
When these are ubiquitous enough, the vast majority of people who currently own cars won't need to. It'll be so much cheaper and easier to use rideshare.
I can't really imagine the circumstances where I wouldn't want to still own my own vehicle even if it had an autonomous mode. I drive it places where I don't have cellular service. I keep lots of stuff in the vehicle. It's customized with accessories like roofracks. I can hop in my vehicle from my house immediately whenever I want to.
If I lived in a city and garaging a car were inconvenient/expensive? Maybe. But that's not me or a lot of other people.
It's not going to be for everybody.
But if it's half the price over the course of a year? And you can summon it in advance cheaply? And it basically never takes more than 5 min to arrive anyways, since they're everywhere?
You might decide it's worth it to keep the stuff you really need in a messenger bag or backpack or something, the way people in NYC do. And maybe roof racks don't matter if you can just summon a second autonomous van behind you to hold whatever you were going to put on your roof.
Obviously if you're a contractor or something you'll need your own vehicle. But the point is that for most people, sure they can't keep stuff in their trunk all the time, but that's a happy tradeoff if the total cost of driving is 50% less.
Of course there will be exceptions or holdouts, but it will come for gig drivers, then for second cars, and go from there. There will be versions with roof racks, with extra luggage space, with child seats.
Statistically, and from a global perspective, the apartment-dwelling car owner (most likely with a lower income than yours) is a heck of a lot more common than living in American-style suburbia or a small town.
I would switch in an instant and get rid of my car
Uber took 14 years to make it to profitability. Money's frequently characterized as impatient, unable to look past the next quarter, but when it wants to be, it can wait.
Waymo's older than Uber, but they hold many key patents by this point. Now that they've started running a taxi service, it seems straightforwards to scale up, assuming that is the business they want to be in. Then it's just a matter of charging more than it costs to run the service, and wait.
What makes investors patient when no profits for years, is when they see growth, entrenched commanding lead and network effects, large user base etc. As long as investors can imagine a good likelihood of eventual profitability, then growth in the present is a fantastic substitute for profits.
Growth tells you the eventual profits will be bigger. Leadership and moat gives certainty that the company will actually get the profits for the market they grew.
Imagine if you could buy your own "Waymo-equipped car". No need for driving lessons. No aggravation. No road rage.
How many people would pay for such a luxury car? With the US population aging and public transit non-existent in most places, Waymo probably has a market for cars.
Assuming there are multiple providers of the software, I’d pay expect to pay normal automotive margins on top of the hardware cost.
That’s probably $1000-2000 per car, or about a penny a mile.
I’m not sure how much a few lidars will cost at scale. The compute board is a few hundred. Modern cars already have plenty of cameras.
There’s clearly a demand for self-driving privately owned vehicles as well, but think of it this way - why own a self-driving Chevy when you could hire a self-driving Cadillac when you need to go somewhere?
I would pay $100k today for a basic EV with Waymo tech. Maybe more. It would essentially be like having a 24/7 personal car and driver available.
I am I really hate driving long road trips.. So yes! Or they could even sell private taxi between states so I don't even have to own a car :)
Tesla has 1/3rd the market cap.
If Waymo is a rounding error to GOOG, it's basically a rounding error to Tesla's implied valuation.
So what is Tesla valued in then?
Clearly not car sales, profit, and especially growth in either of those segments.
xAI is supposed to be where all the AI is.
Where is it?
Future gains on political corruption?
Faith in the fact that Elon has never lost investors money.
Robotics I think.
Tesla’s robots during that investor event were shown to be controlled by humans wearing mimicking hardware
Uber making 50B, probably means Uber is paying drivers around 200B or higher. So that is Waymo’s potential revenue in the long term as it releases in most ride share markets. I think it’s under 1B revenue now, which just shows how much growth ahead is possible. Even if we think Uber will be at least 50% market share in the coming decade, at least 100X growth is left for Waymo. This also completely ignores Waymo creating latent demand, which is wholly possible. I would for example trust a Waymo to drop my kids everyday over an Uber.
Uber also has to pay drivers. How much of that $50B goes to the operator?
Meanwhile, for Waymo, a good chunk of it is profit (after the fixed cost of the vehicle, of course).
The cost of the computers, LIDAR, special maintenance, vandalism, staffing humans for remote issue handling etc will probably costs the same as a year's income for an Uber driver. But after that it's mostly profit and they can run cars longer.
The most important thing for Waymo is scaling up production of LIDAR and maintaining them efficiently. They will have a massive fleet running very sophisticated radar+computers. That's a huge logistical investment when it's a million cars. Those sensors will break or be damaged.
They've been partnering with Uber to maintain the fleet in some cities haven't they since they already have regional infrastructure? I don't think they want to be in the fleet management business.
AFAIK Uber is doing app integrations + some local operational fleet management. Waymo is supplying the cars, radars, computers, remote service, the brand, etc. Waymo has to scale that production and maintenance up country wide and then globally.
Uber's CEO compared it to Marriot, people come in to run the hotels in the local region, but they actually don't own the hotels. It's like hired managers who take a cut.
It also makes sense to have people with local experience run them in each local region. But those businesses still involve margins and expenses that have to make sense.
Don’t forget that Waymo will always be a much lower margin business than search! Setting aside the decades of R&D expense, those cars require purchasing, maintenance, warehousing, etc.
Autonomous cars won't sue you, never sleep, don't go on strike, don't sleep 8 hours a day, keep driving when the car needs obvious repairs.
>Autonomous cars won't sue you
but the companies that own them will or their insurance carriers.
All that may be true. Human drivers are not the point of comparison. The search business is. Waymo will still always be a lower margin business than search for the reasons I enumerated.
Waymo may end up being great business. But it is unlikely to exceed what search is/was. For that reason, press X to doubt GP's claim that Alphabet is undervalued. "IT'S PRICED IN" [1]
[1] https://www.reddit.com/r/wallstreetbets/comments/eberem/ever...
It's unlikely to exceed what search was, but transit is a much more reliable bet for continued revenue. I don't think anyone is betting on ad revenue being reliable at Google long term anymore.
But the market is so, so much bigger. And the margins will likely stay high for a long time while there are few competitors, and their main competition is human drivers.
Not having to pay drivers is an enormous source of profit.
As big as search!? Doubtful. The entire globe is unlikely to be the addressable market. China will never let Waymo in. India will undoubtedly field multiple worthy competitors. Europe is hostile to technological progress and even more-so to American tech cos. In most parts of the world, Waymo is unlikely to be able to deliver a positive gross margin business given the per-capita-income of most places.
It could be a big business. In fact, I hope it is. Lives will be saved. But there is still a lot to be worked out, and the margins will never be as sweet as those of search.
Who pays for search though? Sure it's 100% margin, but it's 100% of not much.
I think the plan is that other entities will own and maintain the cars. That's why they're working with partners like Uber and Avis.
One of the main reasons to vertically integrate is to expand margins by squeezing cost out of the value chain. My point still stands: Waymo will never have margins as good as search.
Indeed. The richest showman that ever lived and successfully duped both liberal and conservative population and politicians. Well deserved I say.
Wild that people will call the founder of SpaceX a "showman"
Let's settle on calling the founder of Hyperloop a "showman".
Yes. Because we should all be judged by our failures.
Hey, Im a fan. Fail fast. Build things.
Most very rich people just sit and roll in their money in the finance markets like scrooge mcduck.
But… I think the performance in the whitehouse was performative nonesense.
What a waste of everyone’s time for the sake of appearances.
More building things, less dancing please Elon.
He has done many impressive things, but one consistent thing about the man is that he always over promises and regularly under delivers. The examples are too numerous to count (smashing the CT's "armour" glass, humans to Mars in 2024, Thai cave submarine, naming your driver assistance technology Full Self Driving, etc, etc)
Perhaps that's simply the price of achievement, but Showman is apt
That is a real, important accomplishment, but he's also a showman.
[flagged]
This is definitely not true and easily observed to be false if you live in the area, then take into account waymo is active in far more areas
You got a soruce
aha - the source is "elon fantasy weekly" :)
Waymo doesn't publish any.
But yeah I didn't realize Waymo's coverage is more than Austin and SF where Tesla rules already. So maybe end of year they'll overtake. Which is crazy Waymo is sitting on this. Even at 10x more expensive cars you'd think they would just put their cars everywhere, but scalability bottleneck seems to be software or lack of remote ops.
Waymo does 250K rides every week
how many does Tesla do? I cannot find a statistic
Don’t forget Zip2, PayPal, Neuralink, OpenAI, and The Boring Company.
There are large swaths of people that accept headlines as fact and/or cannot or will not grapple with nuance and complexity (“I think Elon’s a jerk and he is a formidable engineer.”) Perhaps it’s a sign of these polarized times, or, as I believe, people have always been like this. We just have more time and resources to dedicate to outrage and flamewarring than we did in the past.
I don't deny his accomplishments. On the contrary, I think he is a genius. It's just that he is an extremely, damagingly biased genius.
https://x.com/elonmusk/status/1240754657263144960
Genuine question - are there (or have there been ) any geniuses that are not unhinged?
Yes, there has been nice geniuses (ie. people with extreme talent), Mozart was for example a good person. Da Vinci (if a little sycophantic when young) was not unhinged at all nor abusive and was appreciated.
But since romantism we have built this image of the genius as necessarily abusive.
I’m sure abusive genius are very visible (by definition?) and that abusive people tend to monopolize more ressources too. (Like these tenured professors that use their students to advance their own career)
Einstein, Euler, and Darwin were also nice people by many accounts.
I think you guys shouldn't be comparing “geniuses” because i don't think thats the forcing function here (ie IQ and ability).
The forcing function is having so much responsibility and stress from running so many companies. You have no extra bandwidth for anything. All your time is spent.
So maybe look at comparable people with insane schedules/workloads/very high pressure situations.
Fair. With Elon it feels like there's an obsessiveness that drives him to take on so much responsibility. And as you say, that can affect what he says publicly.
True -- also I wouldn't say Elon is a genius. I feel thats a term for people who solve deep intractable physics/math problems. Elon's admirable attributes are that he is an insane capital allocator, has a very acute engineering mind (rare for leadership), curious mind, sees the future paths, dedicated focus and is an unabashed salesman of his products and philosophy (maybe this one isn't as admirable but its critical to his success).
Euler was famously a genius and a well-tempered family man.
He was ousted from Paypal before anything major happened, he was basically just a shareholder.
The Boring Company is an obvious bust. So is the Hyperloop. Neuralink is another likely bust. Tesla solar is going nowhere. The Cybertruck is a millstone around Tesla's neck. Etc, etc.
He wasn't even a fonder of Tesla. He was just a investor that became the CEO.
And the tweet below makes me question a lot about him. Doesn't sound like a genius to me.
"Lidar and radar reduce safety due to sensor contention. If lidars/radars disagree with cameras, which one wins?
This sensor ambiguity causes increased, not decreased, risk. That’s why Waymos can’t drive on highways.
We turned off the radars in Teslas to increase safety. Cameras ftw."
Yeah, he’s not an engineer. He fools people by regurgitating stuff from actual smart people.
Waymo is driving on the 101 in LA
Tesla’s been doing that for years in SF. There’s only been one fatality on that stretch of the 101 so far.
More info on autopilot deaths (59 including 2 FSD):
https://www.tesladeaths.com/
Waymo’s had one fatality (other driver was at fault), but that’s not normalized by miles driven.
I've driven both SF 101 and LA 101, they are not the same thing.
You can’t compare Tesla and Waymo. Only the latter is truly autonomous.
> That's why Waymo can't drive on highways.
^^^ (they are)
To be fair Tesla was in a very primitive state when he took over.
Usually Elon's technical flaws aren't on display, or at least he covers them well. For example while it's true FSD hasn't worked out, but I don't know you could say at the time "most competent AI devs knew it wouldn't work out". However, when Elon attempted to move PayPal from Linux to Windows, most competent software engineers would have advised against it. Paypal isn't an example of Elon's genius in action - it's the opposite.
When Tesla introduced HW2 it was clear to people in the self-driving industry that it wouldn't work out. Elon was insistent on repeating mistakes that other companies had already learned from. Of course the other companies never considered some people's willingness to pay good money just to pretend that their cars can drive themselves.
> FSD hasn't worked out
Says who? I've tried it and the capabilities are amazing. If you told me 10 years ago that I would be able to buy this in 2025 I wouldn't have believed you.
Says me, who owns a Y that has the FSD package. Random braking on a highway, indicator lights coming on for no apparent reason, windscreen wipers the start on a dry day, attempts move through a red light. None of those things are common, none are serious if your hands are on the wheel and you are giving it your full attention. It's a serviceable attempt at FSD Level 3, and auto park works well.
But when I bought it, Elon was promoting hiring out your car as a FSD level 5 taxi when you weren't using it. If I regularly took my hands off the steering wheel and went for a snooze (if that was possible, which it isn't because they would be sued within an inch of their life), I'd be dead by now.
Yeah HN hivemind is wild. FSD is something you can buy and use RIGHT NOW. Autonomous driving in YOUR HANDS. Waymo still feels like a school project.
Tesla is level 2. Waymo is level 4.
I think the real purpose of the Boring Company and Hyperloop were preventing/slowing expansion of public transit, and that by that measure they were successful.
I think the purpose was to extract money from governments, like most of Elon's businesses.
I don't think it was a carefully calculated conspiracy (such as 1)
I think it was an engineer with found wealth starting to do stuff with it.
but nowadays I think he has evolved into something different, maybe some of it from the wild public feedback loop, some of it because some of the things he cares about are going wildly wrong.
[1] https://en.wikipedia.org/wiki/General_Motors_streetcar_consp...
Elon says it was a conspiracy designed to sabotage high speed rail, just like the one you cited. The Koch brothers helped him:
https://www.fresnobee.com/opinion/editorials/article26445107...
There is nothing in the article, the twitter thread it quotes, or the text from Musk's biography quoted in the respective tweet, that indicates that the Koch brothers assisted Elon Musk in any way in trying to sabotage California's high speed rail. They're simply mentioned as other people that oppose transitioning away from automobiles.
Furthermore, Elon Musk doesn't say that the Hyperloop "was a conspiracy designed to sabotage high speed rail." He is quoted in his biography as saying that he hates high speed rail, doesn't want them to build it, and thinks it's a waste of money. He also says that he had no intention of leading the effort to build Hyperloop himself, where he's directly quoted as saying, "Down the road, I might fund or advise on a Hyperloop project, but right now I can't take my eye off the ball at either SpaceX or Tesla." The biographer speculates that this means it was a cynical ploy to get HSR cancelled, and I don't think it's unreasonable to infer this, but one could just as easily infer that Elon really did want the California legislature to build something akin to a Hyperloop instead.
There's no debating that Elon hates public transit, he'll tell you himself[1]. You don't have to spread misinformation to make that point
[1] https://www.wired.com/story/elon-musk-awkward-dislike-mass-t...
I am "just a shareholder" in Paypal. Elon Musk had a > 10% stake inherited from his ownership of one of the companies that was the precursor to Paypal itself. It's not remotely the same thing. And listing failures is not meaningful at all. Failure is the default outcome in business.
Ever heard of luck? That is my main point.
Wild that people will call a guy who bought SpaceX the founder of SpaceX.
Either go ague with Wikipedia, or put some argument in the comment when making claims you expect people to verify themselves. People are just going to look it up on Wiki.
> SpaceX was founded by Elon Musk in 2002 with a vision of decreasing the costs of space launches, paving the way to a self-sustaining colony on Mars.
https://en.wikipedia.org/wiki/SpaceX 2nd paragraph
Deceiving people doesn't mean you deserve your gains.
"Deserve" is a human construction.
We are humans.
Right, but we're adults, we realize that "deserve" is only as real as the ability to enforce it.
Largely because investors fear that Google's new products (especially AI) will cannibalize its massively lucrative ads business.
Fear is a bit of an understatement
100%. Look at the traffic drop off from google to (insert your fav AI).
Its a real and verifiable threat to their core business. Much larger rev than waymo (current and future).
but if they're google's products how would they cannabalize ads biz. would revenue not just shift? or do you believe ai search will be overly adopted but not as profitable?
I think its the later. And also the fact that they are not the firstmover in AI search. More people know about chatgpt than they know about gemini
Google was late to search, late to smartphones, late to internet email. I'm having a hard time thinking of any of their large markets where they were a first mover, maybe YouTube-ish, widespread user uploaded internet video wasn't meaningfully available before the rise of YouTube.
On topic, Waymo is clearly a first mover in self-driving, having the first legal commercial services.
But, being the first mover is usually more of a disadvantage than an advantage, IMHO.
I'm struggling to think of a single product where the first mover won. At best they are able to hold some market share like Dropbox or Slack, but eventually big tech moves in and crushes them by just offering the same thing but cheaper and more integrated.
Stocks are narrative-driven, and sometimes this aspect swamps the "fundamentals." Keynesian beauty contests all the way down.
But the earnings of Waymo (or hypothetically Tesla) are nothing in Alphabet as a whole.
If you get a great deal on your house and then massively overpay for some avocados, the latter's going to barely move your overall wealth.
I believe TSLA also represents their humanoid robot segment with some questionable addressable market definitions done by investment analysts. I believe it’s overvalued but they are a forcing function for the other tech companies to push ahead
Waymo is a small portion of Alphabets business, while cars are a massive portion of Tesla's. If waymo was seperated out from Alphabet maybe it's p/E would be that high.
PE has been irrelevant since the dotcom crash if not sooner. us equities are no based in reality
Google is just not a risk taker these days. You don't risk you don't get rewarded.
Tesla is literally operating a robotaxi service.
They're operating a Robotaxi service, not a robotaxi service.
If I create a shuttle bus service for my neighborhood and call it the "Space Shuttle", I am not operating a space shuttle.
A whole 15 cars, with "supervisors" in the drivers seat!
And only last week did they even open up the waitlist to non-influencers.
The supervisors are not in the driver's seat.
https://electrek.co/2025/09/03/tesla-moves-robotaxi-safety-m...
The day this news was released, Elon released the video of him talking to the Optimus bot to overshadow the news. Showman gonna showman.
TIL. I stand corrected. Though worth pointing out (as the article does) that on September 1st, new legislation in Texas was passed adding some restrictions to autonomous vehicles. So seems reasonably likely this is more regulatory than necessary.
I wonder what motivated Texas, who was famously very open to such testing, to tighten down the regulations more after time?
They've managed to automate it but reduce the labor costs by zero in the process. Now that's innovation.
Unsafe at any speed
Because some people read beyond headlines and realize that Tesla will most likely dominate with Robotaxi. Their traditional consumer vehicle revenue could pale in comparison. And Optimus could be another order of magnitude larger.
That’s the optimistic bull case. It’s not impossible.
Tesla will be able to scale Robotaxi much quicker than Waymo can scale.
Why? In principle the basic Waymo technology could be adapted to work on any modern vehicle. They aren't dependent on Jaguar manufacturing capacity to scale up.
It's capital intensive to make all of those devices. Tesla's strategy is to rent back devices they sell to consumers. This lowers the necessary capital costs and will enable quick scaling. It's a similar ploy to how Amazon quickly grew its delivery capabilities.
Alphabet has $95B of cash and short-term investments on hand. I don't think lack of capital is the obstacle to scaling here.
https://abc.xyz/investor/sec-filings/quarterly-filings/2025/
Tesla still has no autonomous vehicle that customers can actually buy, let alone rent back for taxi service. So any "strategy" remains entirely hypothetical.
>Alphabet has $95B of cash and short-term investments
Not only that, but also they could probably raise 10 times that much by creating new shares and selling them (if they had a plausible story to tell investors as to why the money would be well spent).
I so wish Tesla had gone Waymo’s route and focused on delivering really safe LIDAR-based level 4 with cheaper hardware. I think they’d be well-positioned to take the market by storm.
But instead they made an ideological stand on cameras only, and they’re helmed by an unhinged drug addict who lies constantly, to the point many who once would have loved to buy an actual self-driving Tesla now won’t touch them with a ten-foot pole, even if they do someday deliver an L4 experience.
I really, really hope Waymo licenses their tech. I think that would stomp Tesla into the ground once and for all.
it's bizarre how even the smart people here (smarter than me on average, no doubt), think they know better than Karpathy and Musk who have spent a decade deep in this problem.
SpaceX has landed orbital boosters 500 times, and STILL no one else has done it.
Teslas drive millions of miles autonomously (yes supervised, but still) every day.
You can't even type a response without containig your political/social bias on anything related to Musk.
Tesla is still at level 2 autonomy, whereas Waymo is at level 4.
It doesn't appear Tesla can achieve level 3 autonomy given Musk's ideological opposition to lidar. Without lidar, the AI has to be 100% accurate, and it's not and won't be for the foreseeable future.
As they say in Maine, "You can't get there from here."
The level 2 vs 4 thing is a technicality and fairly misleading, Teslas can already operate in a much wider range of roads and conditions. Waymo is still on guardrails. Mercedes touted "level 4" but you had to be following another car, going slow, etc etc. General autonomy is what matters.
I trust Karpathy when it comes to lidar vs vision. Do you shoot lasers out of your head to drive?
LIDAR sensors are getting cheaper faster than camera-based autonomous driving software is improving. I predict that in a few years, regular luxury cars that are still mostly human driven will come with LIDAR for collision avoidance and improved driver situational awareness. Just like they already use RADAR for the same purpose today.
Ah yes $100k worth of extremely complex sensors. very scalable.
Not a problem. The costs for sensors always falls rapidly as production volumes scale up. The first GPS receivers were large dedicated devices that cost thousands of dollars. Now they're just a chunk of IP in the SOC for every little consumer device and cost pennies.
There are about 1,500 Waymo cars in existence, versus about 7,000,000 Teslas in the last seven years.
But there are 0 Teslas that are as effective at self-driving as Waymo, so they're still ahead.
My Model Y in Vancouver drives me to and from work daily. I cannot get a Waymo here -- and I certainly cannot purchase one privately. Which is more effective where I live?
Teslas have a ~about 500 miles between interventions (they don't release actual data, no surprise), whereas Waymo is at around 17,000 miles.
That's a 34x divide. At full scale that's something like 30% of Teslas having an intervention every day.
I don’t doubt that Waymo car is more advanced than FSD, but that comparison isn’t as impressive as it sounds. The numbers of FSD equipped Teslas dwarfs that of Waymos, and they are available everywhere, not just selective cities. You have to take that into account.
Teslas is also much cheaper, and easier to scale. Tesla has better growth potential even if their tech is less impressive.
It's not that their FSD tech isn't less impressive, it's that it's not FSD tech.
Even worse (for Tesla) is that if they do try an make their non-FSD tech do FSD, and it decks little jimmy because the flashlight in his hand looked like a far off street light, Tesla is liable to face a knee-jerk federal law mandating lidar. And just like that the dream is dead.
This forces Tesla to be extremely paranoid, as it's one visual mistake away from being told to use lidar.
Why is a 34x improvement in the rate of interventions not as impressive as it sounds?
I’m not even sure that Waymo number is still correct. They’re doing hundreds of thousands of paid rides per week, with no one in the front seat, so not sure what an “intervention” even means at that point. Maybe where the passenger needed help and called support? That’s 1000x better than needing to grab the wheel because your Tesla was about to drive into oncoming traffic or run over a kid in a wheelchair.
You are supposed to supervise Tesla FSD. Waymo doesn't require someone in the driver's seat at all. They aren't the same thing.
We’ve also not seen how capable Tesla is at evasive maneuvers. We have plenty of videos (hundreds now) of Waymo making instant swerves to avoid children running onto the road, cars running red lights, a person falling from a Scooty etc. These are not maneuvers you would expect from a human, which shows how Waymo has pretty successfully crossed the human bar in safety. If Tesla does not demonstrate this, on top of driving normally, I don’t think they have a product. The barrier to give control to a computer is super human not human like driving.
Also philosophically I don’t see how a big neural network will create such evasive maneuvers, unless you try to create such scenarios in a simulator and collect evasive data. Seems prohibitively expensive to do so in the real world.
Market says “as effective” doesn’t matter. Needs to be “good enough”.
I mean FSD is pretty good and useful. But yes, not unsupervised.
The Coca-Cola company sells even more units than Tesla, but if those units don't drive themselves they're moot to this discussion.
Same could be said about Tesla when it started.
Overvalued by traditional (PE) means. I've ridden in Waymo (50+) and Austin Robotaxis (12). Tesla has Waymo beat in terms of human-like feel, interior features (sync to your own Spotify, Youtube, etc). When Tesla removes the passenger seat monitor, scaling will happen much faster than Waymo... Tesla just received the initial license for driverless Robotaxi in Nevada. Tesla also produces more Robotaxi-capable Model Ys in ~6 hours as Waymo has cars in service (in total).
Tesla's self-driving technology is a joke compared to Waymo's and the Tesla brand is extremely toxic now. I see from your other comments that you're big on Tesla (own several and have a son who works there) but as an unbiased observer I cannot fathom them winning this market.
I have 2 AI4 Teslas with FSD, and I don't find V13.2.9 lacking at all in the Vancouver area. V14 will be a 10x increase in parameters, too. Why do you feel it's a "joke"?
It's a "joke" (I wouldn't call it that, but it's a vastly different product) because you have to pay attention to the road at all times.
You don't live in a Waymo city, so I understand. A lot of people who don't live in a Waymo city don't really get it.
Waymo is a completely different product than FSD. It's a robot that comes and drives you from point A to point B. You can do whatever you want while it's driving, such as take a nap or work on your laptop.
> I have 2 AI4 Teslas with FSD
That is false. No Tesla is capable of full self-driving.
Mandatory supervision by a human on the driver’s seat is not full self-driving, no matter how much Elon insists on calling it that.
Tesla was SAE level 2 in 2013, and they are still SAE level 2. Waymo's Robotaxis are SAE level 4, and they can drive on public roads empty with no human supervision, both technically and legally.
Also safely.
> When Tesla removes the passenger seat monitor
This is a huge jump, possibly still 5+ years away.
I have friends on the Autopilot team (and a son). Their goal is by end of year. I've been on HN for 15+ years, and seemingly the only downvotes I get are when I post my thoughts and opinions on Tesla.
Tesla FSD has been autonomous by the end of the year for 8+ years now. Don't believe people desperate to make Elon's lies seem plausible.
It would be great if Tesla figured out how to do safe autonomous driving with their very limited sensor suite, but there's a lot of reason to be skeptical: https://en.wikipedia.org/wiki/List_of_predictions_for_autono...
I wouldn't not be surprised if they figure out some very narrow way to have no safety driver in the car (1:1 remote ops?) by the end of the year.
1:1 is going to be ruinously expensive. You need three shifts of remote operators. Even in the Philippines or Vietnam, if you can make the latency work, that's prohibitive.
> 1:1 is going to be ruinously expensive.
I agree, but this is how taxis/Ubers work.
How do Elon Musk's predictions relate to Tesla achieving a robotaxi service or not?
Ignore his predictions and just... look at whether or not the Tesla FSD team is making progress.
> How do Elon Musk's predictions relate to Tesla achieving a robotaxi service or not? > > Ignore his predictions and just... look at whether or not the Tesla FSD team is making progress.
I'm seriously baffled by this comment. How can Elons comments not be relevant? How are you proposing we assess the progress of the FSD team? And why should the assessment be different to the last 5 years where FSD was supposedly ready (according to someone with intimate insight into the work of the FSD team) by the end of the year?
> How are you proposing we assess the progress of the FSD team?
...any metric you want? Miles driven under FSD. Miles driven without intervention. Miles driven without accident. Anecdata from friends of yours who own a Tesla. Whether or not a partially supervised pilot program has been launched in some cities.
If Elon Musk said in 1999 "I think we will achieve self-driving next year", that also has no bearing on whether or not self-driving is achieved in 2025 (in either the positive or negative direction). It only means that Elon Musk's "predictions" can't be trusted as an accurate harbinger of success. Which is precisely why you look beyond his words and at the reality on the ground, which strongly indicates Tesla has made a huge amount of progress in the last 10 years, and could be very close to having unsupervised robotaxi service in various jurisdictions.
If we use kilometers driven with drive assist as a metric then nearly car manufacturers will have robotaxis by the end of the year.
If we talk about anecdotal evidence then I know people who are deeply familiar with the topic (working of self driving technology at other manufacturers) and they say fully self driving is still many years away for all manufacturers. Moreover the general industry sentiment is that Tesla is behind now and that more sensors then just cameras are needed.
But instead I should believe the Tesla fan boys who just like Musk have been raving about the amazing progress and telling me that FSD is just around the corner for years.
> If we use kilometers driven with drive assist as a metric then nearly car manufacturers will have robotaxis by the end of the year.
Sure, if you pretend that highway lane-keeping and universal A-to-B navigation are the same thing.
"What competitors say" is quite possibly the worst anecdata you could find as a broad rule, no? There is a wide gap between that and "Tesla fan boys".
Never heard of "universal A-to-B navigation", that sounds like google maps.
Is it fully self-driving, like Waymo? If not, then I'd lump it in with anything else that isn't fully self-driving. Either I can safely and legally nap while commuting or I cannot. Something that requires me to actively supervise the car and intervene as necessary is not self-driving, it is drive assist.
> "What competitors say" is quite possibly the worst anecdata you could find as a broad rule, no?
The post you're responding to is not simply repeating what competitors say, it is speaking of using data to avoid trusting what anybody says. Thus, this isn't a fair comparison. It should also be noted that you yourself suggested that the poster use anecdata.
That said, what tesla says about themselves is even worse than what tesla competitors say, if only because tesla is infamously untrustworthy, and their competitors are not.
But again: don't listen to what tesla et al say they will someday do, compare the data for tesla's drive assist vs tesla competitors' drive assist.
FSD is widely considered to be off its originally-stated goal by at least 5 to 6 years.
Can we expect you to come back on Jan 1, 2026 and provide an update?
> Their goal is by end of year.
It's like what 6-7 years since the goal was "end of the year".
You're not "posting your opinions on Tesla", you're literally shoveling them into everybody's throats. You'd be "posting your opinions" when it was one, two comments, and not plenty, like under this news. You're a Tesla freak or fanboy, not an objective commenter.
Yeah, I also heard Sky Ferreira’s album is coming out this year.
> Their goal is by end of year.
Ummm.
I think the downvotes might be due to one or more of the following:
- You're uncritically parroting the notoriously untrustworthy talking points of a notoriously untrustworthy company, and HN posters expect more critical thought in comments.
- You're redirecting to some rumored "goal" rather than a realistic prediction, which was the topic, and HN posters liked the topic.
- HN posters may think that your vested interest in tesla behooves you to think more critically than the average person on matters involving tesla, rather than less, to overcome any implicit bias you might have.
- I have a goal of end-of-month, so that means I'll have it even sooner than tesla, right? This is how many view the claim by tesla, except I, a random person, literally have less of a reputation for dissembling and failure to deliver than tesla does.
Waymo does not have YouTube sync, but they do have Apotify sync.
> When Tesla removes the passenger seat monitor
They literally moved that monitor to the driver's seat! Progress, indeed.
I drove for Uber/Lyft back in 2020 and let me tell you, SFO is a nightmare. I missed a turn once and had a passenger trying to make a flight furious at me. I quickly figured out there were a group of drivers who specialized in SFO and amatuers like me should avoid the place. When Waymo announced San Jose I thought ok, that makes sense because SJC is easy, but SFO? Wow, I'm impressed. I hope it goes to plan.
Those turn offs for specific terminals are very small and easy to miss.
Nothing more rewarding than a company working hard and seeing real-world, first of its kind results in action. Makes me feel giddy about a company again like peak tech back in the 2010 era.
Congrats to the Waymo team, I’m sure this was a huge milestone internally.
Looks like this Kiss & Fly area where pickup will be is at the car rental center.
Oh, this makes a bit of sense. The Avis/Budget fleet team will be part of managing the vehicles, so they can be quickly cleaned and fueled up when they slide into the airport, too.
https://www.avisbudgetgroup.com/avis-budget-group-announces-...
(Dallas, but they do this in other cities, too.)
huh. I didn't even know this existed.
Same. I go to the rectal car center at least 4 times each year. I just was there on Saturday and had no idea either. Still don't know what it is other than Waymo pickup.
>rectal car center
Known nickname or typo?
Definitely phone autocorrect issue. I'm gonna leave it though
How often do you type "rectal" for that to become an autocorrect default for you??
If you are over 50, and serious about not getting colon cancer, maybe a little bit more than one would expect.
OP is a urologist
I use Google keyboard without customized auto correct.
It really likes to change random words to inappropriate things.
But I guess that's the people who are typing on phones a lot are typing about.
Otherwise known by its popular name “Cloaca-Rent—A-Car”
Or both :-D
It - along with cell phone waiting lots - are ways for people to drop others off and avoid the traffic around the terminals themselves.
Which can be bad - I often find it easier to just pay for a few minutes parking on dropoff/pickup.
Yeah, but the SFO cell phone lot is particularly bad, and traffic to the terminals is bad.
On the flip side, there are airports like Cleveland where people just park their car at arrivals and disappear for 20 minutes.
I did always find the term kiss and fly confusing and weirdly intimate, as if everyone is getting a ride to the airport from a spouse or parent. Definitely a throwback to another era.
I think it's also a regional thing; I'd never heard of it.
ChatGPT tells me it originated at Paris CDG in the 1990s where it was "depose minute" but then also cites LA Times from the 1950s as the first source of the term[1]. I've seen it in Germany as "Kiss und Ride".
So it's not really a regionalism, but I also don't think it's super common.
1. https://en.wikipedia.org/wiki/Park_and_ride?utm_source=chatg...
> rectal car center
That's way mo' information than needed thanks.
But seriously. I wonder why they have a designated pickup point if it would make sense to spread the cars out to alleviate traffic bottlenecks.
What's even better is the variety of names this thing has. I'm my area, it's the "cell phone lane"
Does this mean they'll be able to take the freeways to get there? Surface streets from SF to SFO would be pretty slow.
I'd hope so. As an aside, I wish Waymo was more transparent on the app that their cars are not allowed to take passengers on the freeway. I was unaware of this restriction when I booked a ride from SF to Burlingame last month and I was stuck in a Waymo for an hour going down residential streets!
Doesnt it show the route and the ETA before your book the ride?
No, they need to add a pop up with even more text users will not read.
They have had permission to be on freeways for a while [0], although so far they have only done employee testing (I believe)
[0] https://sfstandard.com/2024/03/01/waymo-san-francisco-cpuc-e...
I wonder how that'd feel. I took a Waymo in SF last fall and I was pretty impressed. But it was also slow city speeds. I wonder if it feels different going at freeway speeds with "no one" at the wheel.
While the margin of error is much lower on a freeway due to the speeds, other drivers are generally a lot more predictable (also in part due to the speeds).
Sure - a good freeway is actually a lot more predictable in most circumstances than city driving, so as a problem to solve it's likely a little bit less complicated. What I wonder about is what it feels like as a passenger. I wonder if it would be more or less frightening than being a passenger when my 17 year old is driving.
I use adaptive cruise control a lot, where I rely on the car for keeping a safe distance.
I have a limited version of SuperCruise which means it operates hands-free on freeways but nowhere else. My wife's Equinox EV has the regular version, which operates on a lot of arterials near us and has more capabilities. The first time that the Equinox signaled, changed lanes to pass, signaled, then changed lanes back was shocking.
We moved to a small town and drive a lot more than we used to and I find that having those capabilities really helps relieve the stress.
I will say that I move to the center lane when going through a notorious set of curves on I-5 in Portland because my Bolt doesn't steer as smoothly as I'd like near the concrete barricades. I wanted SuperCruise because it has a fantastic safety record. There are lots of times it's not available but when it is, I have near-total confidence in it.
I took a Waymo that drove on an 'expressway' which had a speed limit of 40mph and it was definitely a different feeling. I did feel a bit scared, at 25mph it feels like a gentle theme park ride, at 40mph it's beyond that and feels dangerous.
There was a good overview on here a while ago about the challenges[1]. You need to plan longer in the future and your sensors need to reach further. It's also a much bigger challenge to collect sensor data as fewer diversions happen per mile (but those that do have higher stakes).
[1] https://news.ycombinator.com/item?id=38973404
Roads that get used more collect more debris. They also break and require maintenance more often. That maintenance is exceptionally disruptive to the normal operation of the road.
Other drivers aren't your only challenge out there.
From the article “ Pickups and dropoffs will initially start at SFO’s Kiss & Fly area – a short AirTrain ride from the terminals – with the intention to explore other locations at the airport in the future.”
They're still working on freeways, doing employee riding testing.
Waymo already had the permit, but they're just being (overly) cautious.
The surface street route that bypasses 101 near Brisbane is surprisingly often faster than 101.
People love crashing there.
Hopefully Waymo does a better job than SF Uber drivers. I can't tell you how many times I've had drivers make a wrong turn trying to find their way to the pickup point.
Boy, if they could actually navigate terminal traffic, I’d give ‘em true self driving.
SFO traffic is not bad at all. Send them to LAX and we're talking.
Its not true self driving until the Waymo asks if dropping you off at arrivals is ok.
There's the big sign there telling you to go to arrivals for drop-off. This is probably a stupid question but can Waymo cars interpret those temporary display signs and follow them? Would it?
It seems to handle the standardized ones (think "construction ahead, detour left") perfectly well from the rides I've taken, but there's all sorts of ways they could be 'cheating' on that.
Thats AGI
I'll be honest, I think LAX's traffic is better than SFO's. It feels like there's a lot less spaghetti at LAX, the shortcuts are reasonable, and you don't have separate international and domestic loops.
LAX's many parking lots with left lane entrances definitely threw me for a loop the first couple of times.
Overall though, I think I agree with you.
JFK is probably the 10th circle of hell
I was also thinking JFK is pretty bad.
I don't touch JFK with 10mi pole. I've always found EWR to be much more consistent and easier to get to
Send them to BOS and we're talking
Is the mark of intelligence being able to navigate to BOS, or refusing to drive through the big dig in the first place?
The Big Dig, for all the digs it rightfully got for taking forever and costing a shitton, actually does the job it's supposed to (mostly). I'm generally pleasantly surprised how few problems it had when I lived there.
I didn't go to Logan a ton though.
I haven't driven through it much since it was completed - I moved out of Boston in 2005. As a now-just-visitor I find driving through the central artery tunnels even more stressful than on the old elevated 93, but that could just be a lack of familiarity thing -- I don't drive when I visit Boston, only if I'm flying in to Logan to head to the cape or something. It is, however, _really_ nice not having the freeway cutting through and the greenway is gorgeous.
JFK has entered the chat
I'll take JFK over LAX. The construction going on right now at JFK sucks, but LAX is comically bad. Just last week I was on a rental car shuttle at LAX and watched 3 separate groups of people at different terminals miss their flights because traffic just wasn't moving.
Maybe, but this approvals only allows them to go to the rental car center, which is quite far from the terminals. The passenger will need to take the air train to the terminals.
They already do in PHX.
If Waymo can pull off airport pickups smoothly, it might shift how we think about edge entry to city traffic. Most cities still struggle with that "last mile" problem maybe self driving fits perfectly there.
One concern I have is how the user data collected by self driving cars will be handled. Companies like Waymo likely hold even more data than Uber. If such data is truly used in sensitive locations like airports, I hope there will be clear and transparent mechanisms.
Seems like Tesla keeps talking big, while waymo conquers city by city.
LOL
Can you handle parking structures? I heard a lot of the autonomous cars were using 2D maps and couldn't handle multiple levels. Haha! This was just a year or two ago.
Google maps has been able to figure out parking structures for me recently. Not sure what technology is involved (gps isn't great for vertical) but it's clearly possible.
Google has been collecting data on building interiors for several years now. Not just parking structures. This data is currently used in streetview. Google's geospatial data is unequaled and maybe a bigger advantage than is readily visible.
Do they need a "map" of a parking tower though, just like how humans don't exactly need Google Maps inside of one? I feel like this is something self driving + vision (exit signs and arrows) can handle
one of the waymo depots in SF is a multi level parking building
Do taxis need to park tho?
I mean, depending on the situation, of course. Do taxi drivers in US drop people right in the middle of a busy street?
Kinda yeah? They certainly don't navigate into long-term multi-storey parking structures.
Fair point, both extremes don't really happen in reality.
hilariously needlessly complex expensive solution because "anything but viable public transport!"
There is viable public transit from SFO to the city. Both Caltrain and Bart have decent connections to the airport.
Practically, travelers with heavy check-in bags prefer taxis. Public transit can't be the answer for everything.
Without heavy check-in, transit is a great choice. Alas, some problems remain. Many residents consider BART unsafe. Frequency is pathetic for a major subway line. BART is unusually expensive ($14! - 6$ + 6$ + 2$??) and last mile connectivity is bad.
On the topic of cost, other airports that are on the transit corridor (LaGuardia NYC, Boston Logan, Ronald Reagan Washington) cost $3-4 for the trip. EWR & JFK are more expensive, but they're often faster than taking a taxi which allows them to compete on pricing.
I think this is a really exciting development for Waymo and the future of autonomous transportation. Imagine the next steps; waymo flights
Man, I’m flying into SFO next week. Wish this was already available; I’d take it in a heartbeat.
Waymo ride costs are getting really expensive in SF.
Not sure if you have a recent side-by-side example with Uber, but this seems like it would have to happen if the demand is there. How else can you offer a quality product (i.e., car shows up in a reasonable amount of time) if you don't have enough cars to satisfy the demand? Pricing is the primary demand lever.
There's so much polarizing opinion on Tesla's offering and whether they'll get to Waymo's level sooner than later, but this seems like it's going to be or already is a huge issue for Waymo where they can't manufacture the vehicles fast enough to satisfy the demand as they expand both locally (because they capture more of the market) and into new geographies. Will they try and acquire a manufacturer? I don't think that's economically feasible for Waymo (Geely market cap is $25b, per Google snippet fwiw), and obviously being in the car business is different than autonomous, but I'm sure Google would bankroll a purchase if they thought it was the right growth strategy.
I guess Tesla, even if their autonomous is on par with Waymo tomorrow, also has to manufacture the fleet, but it seems extremely beneficial to have that capacity in house vs. relying on partners. Maybe I'm wrong and it's not that much of an advantage, but at first glance it would seem to be.
The CEO of Uber was quoted as saying Waymos complete more rides per day than 99% of Uber drivers. He didn't give a precise ratio but this makes me think that hundreds of Waymos can replace thousands of Uber drivers and their cars.
CMs like Magna have the flexibility to manufacture, at the low end, hundreds of vehicles, and at the high end thousands. I doubt Waymo will ever make their own vehicles. They are already working with Toyota on adapting Waymo technology to privately owned cars. That implies mass production. That would be a supply of vehicles that are probably simple to adapt to robotaxi use.
That's a crazy statistic and an interesting one for him to actually say out loud. Was that in the context of Uber partnering with Waymo in Austin? And thanks for the insight on the manufacturing side. Sounds like it might actually be to their advantage to use third parties because you can spread the demand around and since auto margins are not high the added cost for that benefit is minimal.
Private cars have a ridiculously low duty cycle. They mainly sit around waiting for their owners to use them. I suppose at some point in the future there might be a traffic jam of autonomous vehicles, but only if the providers are antisocial and don't coordinate ride destinations and routes.
@harmmonica I do. we prefer to use Waymos in SF but Uber has been a lot cheaper in the last six months or so regardless of time of day... Also saw some Zoox self driving boxes on the las vegas strip last week but no one seemed to be using them.
Thanks for sharing, oliver. For anything local I've almost entirely switched and I guess I haven't been doing much price comparison between the two. One thing I have noticed here in LA, albeit only a couple of times, is that during rush hour the waiting time is significantly longer for Waymo. I've taken some Ubers because the wait for a Waymo has been way too long.
Zoox just started allowing rides in Vegas 6 days ago.
I've seen them driving around SF as well, but they're not yet available here.
Waymo is a premium ride product that happens to be self driving.
I think self-driving is what makes Waymo premium. A Waymo is a better experience because the Waymo is safer and smoother driver than any human could ever be.
just use Robotaxi. 1/3 of the price, sometimes less
How much per mile? For some recent example rides, let's say. One I took was exactly $2 / mile but not in SF.
What's that? An app? I see a Chinese app of that name in the android play store, but it only has about 1k downloads
Only for iPhone at the moment
https://apps.apple.com/us/app/tesla-robotaxi/id6744257048
The Tesla service is colloquially called "Robotaxi".
Not just colloquially but officially...
https://www.tesla.com/robotaxi
For now...
Not sure why you're downvoted. I've tried Robotaxi a few times and has been great. They still have a safety driver these days and wait time is a big high though.
I wonder what the ultimate price of this service will be compared to alternatives.
It will remain higher for a while. From reporting I have seen, they are close to maxing out their vehicles, and many people prefer it to other options, so are willing to pay a premium. As long as that is true, it's going to be priced as a premium product. It won't be until fleets grow significantly in size and/or another driverless taxi service enters the market that we will maybe start to see prices driven down closer to marginal cost of a ride.
-edit- multiple other comments apparently disagree with this. I'll defer to people who actually use them over the reporting. Odd that there is that disconnect though.
I know this is only a single data point, but I recently took one in Hollywood. Uber Lyft quoted $33 and Waymo was $20
It's also higher right now because it is a novely. Plenty of people are booking it just to say they rode in a Waymo and take pictures. When that wears off they will have to start competing strictily on price and wait/ride time.
Lots of people, myself included, pay a premium to not have a human at the steering wheel; it's nice to have the car to yourself.
Yourself and three dozen recording devices and call centers full of people tracking the car and reviewing the footage, yes.
To be fair, Waymo claims to not record or transmit audio without you either manually engaging such (by requesting support), or a very unambiguous announcement (presumably when the car gets into some sort of emergency situation). And lying about that claim would probably run afoul of California's 2 party consent law. So still a step up in privacy versus having someone in the car listening in on your conversation.
That said, even if they were listening to you, there's a lot of things that are completely inconsequential from a perspective of an anonymous call center employee far away listening in on, that I probably wouldn't want to talk about in front of a taxi driver.
I know this is somewhat besides the point of the discussion, but.. many Ubers have recording devices inside the car too. The drivers have gotten savvy and protect themselves from false claims or even harassment.
I still count that as a win.
Like, the driver's presence bothers you? Even if they don't talk?
This is just me, but maybe helps explain it. It's not that the presence of a driver is bothersome, but in the pre-Waymo world your interaction with the outside world starts when you step out the door of your house. Now the interaction with the outside world starts when you get to your destination and step out of the Waymo. I really enjoy the outside world, mind you. But it just feels easier to traverse my local area in solitude and with a consistent and comfortable vehicle, and non-erratic driving style.
I imagine how nice it could (will?) be when you can hop into a self-driving car for a longer ride or even a road trip. I think you'll feel like it's an extension of your living room vs. being in a car.
Christ that's depressing. A 'perfect' world where we never interact with anyone.
We need to stop normalising mental diseases
By your logic, we would never have normalized running elevators without an operator.
If you step back do you really think that's indicative of a mental disease? Does it make any difference to you that many times I'm taking a Waymo to go and hang out with friends? Not much of a stretch to say it's allowing me to socialize more because I don't have to worry about my meter running dry, or having one too many drinks to drive myself home, or being able to move around from area to area in comfort. And if you say "you can do that with an Uber too!" it's true! But does it really surprise you that someone would want a car that drives calmly, obeys all traffic laws and gives you a little downtime from the outside world pre or post the activity you'd been doing before stepping in the car? Does that really rise to the level of mental disease?
It seems like a huge catastrophizing stretch to get there based solely on preferring to be in a Waymo rather than a taxi or Uber.
edit: grammar
Well on second reading your comment reads like an ad, given advertising isn't natural either, you can understand the confusion
But I was referring to the wanting the outside world to resemble your house and to have little interaction with humans. No, that's not normal, despite any sophistry or ad speak
The problem isn't when they don't talk and just drive, the problem is when it's late at night and the passenger is a woman who is inebriated. Not having a driver entirely makes that much harder.
Yeah they need scaling and competition before the prices get lower. As long as supply is saturated with demand and nobody else is on their level, there's little reason to lower prices.
Yeah, and just to add even though it's implied in your comment, there's plenty of reason to keep prices where they are independent of a desire to increase revenue. Customers will not wait forever for the car and so if the demand is high you have to keep the price high to discourage people from using it so wait times remain in check. Tricky tightrope they're going to be walking while they optimize the fleet size for local adoption and geographic expansion.
On other threads I've seen conflicting anecdata regarding pricing being higher or lower than an Uber ride. That's not too surprising since the supply and demand variables are going to be different for Waymo.
Their goal is to have lower cost Hyundai models hit the market though, right? So the Jags probably remain the premium/higher cost option.
In my experience so far, Waymo costs about the same as an Uber when you take into account tipping, but takes longer (they're not yet doing freeways). With the addition of SFO to their zone, I can't imagine freeways are far behind, because getting from the city to SFO without using the freeways would be... a novelty.
That's not been my experience... 90% of the time when I check, Waymo is still a good 20-50% more expensive in SF, when comparing to a tip-included Uber or Lyft price.
I've used Waymo countless times in SF. It's typically 15% cheaper than an Uber/Lyft and trip time/wait are generally the same. I much prefer the Waymo.
I've never encountered it being cheaper, what hours do you generally use it?
Generally between 11a and 7p. Going to lunch/dinner.
Self driving taxis are fundamentally much cheaper to provide.
- No driver to pay. - Smaller simpler car. - Can drive 24h a day. - Needs much less parking space.
But fully realizing these benefits is probably a decade away.
Cheaper than uber rn. Long term once they own the market? Too much.
They still have to compete with alternate modes of transportation such as buses, bikes, trains, e-scooter rentals, self-owned cars, Uber with human drivers.
If it would be "too much", then there's no reason why taxis (incl uber/lyft) wouldn't be too much today.
I don’t really think they have to compete much.
Direct competitors are uber and Lyft which they can undercut since they don’t pay drivers.
The people who want to take buses and trains will continue to do so although Waymo might sway some with their ease and if pricing is reasonable.
Bikes and e-scooters only get you so far. Last time I was in SF I didn’t see too many bikes but I saw a ton of e-scooters. Are you really taking an e-scooter further than a few blocks? And when it rains?
Self owned cars make sense for longer trips out of the city but parking is a pain and driving is stressful so this is an easy win for Waymo.
It’s cheaper now so they can take market share. And their cost will certainly be lower than Ubers so they can win the pricing battle. But long term monopoly gonna monopoly. Perfect pricing is a given with the wealth in SF and how many rides will be on a business CC.
from what I heard, the intention is to make it much more affordable than it is now. I don't remember the source right now but I did think it was a blog post or something like that.
I think if it's affordable then people will easily take that. instead of drinking and driving at night or other unsafe activities. if it's affordable then people can just take a waymo home and then back again to get their car when it's safe again.
Certainly they aim to make it affordable now in order to undercut Lyft and uber. Long term they will own the market and jack up prices as monopolies do.
That's great to hear
The title makes it sound like GA but it's still in testing
But still no Waymo app in the European App Store for iOS.
Waymo has planes too??
at first I thought they were doing those cargo quad copter things...
Looks like this would (eventually) include service to not just San Francisco, but also the Peninsula (Silicon Valley) via freeways.
I certainly hope so. And if yes, I imagine people are gonna start using it as a transfer point, to take a Waymo from the peninsula to SF.
Is Waymo L5?
L5 means the car can drive everywhere a human can. Waymo's refuse to drive outside of a constrained area, and occasionally stop to ask for assistance, so that makes them L4.
This whole autonomous driving levels kinda muddies the waters. Some would argue this isn't full L4 even. But it is a self driving car in the places it offers its services.
What would be the argument that Waymo is anything except L4?
No, L5 is a car that can drive itself anywhere in any conditions.
I think there's an implicit "where a decent human driver could drive safely" for L5, otherwise you get increasingly ridiculous scenarios like, "can Waymo drive safely in a whiteout blizzard?" or "can Waymo safely escape an erupting volcano??"
Too bad waymo is more expensive than uber most of the time
Why is that too bad?
is waymo really that good???
how good it compared to Tesla FSD/Robotaxi ???
haven ridden in both a few times, yes, Waymo is head and shoulders better. It's smooth and I don't think I've ever seen any false alarms or behavior that made me feel unsafe in a Waymo, while I've had a few scary or annoying situations in the Teslas. I took a 6-minute robotaxi in drizzling weather where it parked in intersections twice because the cameras were obscured. Meanwhile Waymo can drive perfectly in heavy fog.
Both the Waymos and Teslas have that central display that shows you what the car sees (pedestrians, dogs, traffic cones, other cars, etc). The Waymo representation of the world reaches pretty far is is pretty much perfect from what I've seen. Meanwhile the Tesla one until recently had objects popping in and out.
Neither is perfect, of course; both will hesitate sometimes and creep along when (IMO) they should commit. But they're both still way better in that regard compared to the zoox autonomous cars I see in SF.
Tesla doesn't have a real robotaxi yet, they're still in the testing/prototyping phase where they need a safety driver or safety monitor in the car.
They might be close to a real robotaxi in some areas, but it's hard to say until they actually pull the trigger on removing any employees from the car.
Waymo cannot scale. So for most people it's irrelevant.
Tesla FSD makes driving 90% less taxing mentally. It does 99.9% of the driving perfectly. And its getting better. We are quickly approaching a situation where people who don't drive Teslas are like people who cut their grass with Sickle as compared to people who have driving lawn mowers
There are many cars on the market with ADAS. Tesla is competing with those companies, not with Waymo, which operates in a completely different market.
https://cars.usnews.com/cars-trucks/advice/cars-that-are-alm...
tesla robotaxi is worse than waymo was 3 years ago when I was a tester
Until just now I had no idea Tesla had a taxi service. Otoh I've seen hundreds of Waymos in SF and the west side of LA.
An interesting thing about this is that there are fewer than 1000 Waymos in the SF service area. I don't know today's total, but I'm pretty certain that there are fewer than 5000 Waymos in existence. Maybe as few as half that.
Some months ago Waymo claimed to be providing 250,000 rides per week. If the fleet size was 2500 at the time, that would be 100 rides per vehicle per week.
Wait, what is special about driving to/from airports?
What’s special about the airport is that the City of San Francisco owns and regulates it (as opposed to the streets that are regulated by the state CPUC), and the Board of Supervisors previously were regulatory captured by taxi medallion owners and Teamsters union (https://missionlocal.org/2024/12/waymo-rolls-toward-san-fran...). Specifically, Aaron Peskin (BoS supervisor from 2001–2009, 2015–2025, and board president for the last 2 years) said, “Their entire M.O. is, ‘The state regulates us; we don’t have to work with you, we don’t have to partner with you.’ My response is: There are things we do control. Including where you charge your cars. And the airport. What I intend to do, is condition their deployment and use of the airport property on their meeting a number of conditions around meeting this city’s minimum standards for public safety and transit.” https://missionlocal.org/2023/11/waymo-rebuffed-by-sfo-sf-gu...
I’d say it puts a lot of Uber (and similar) drivers at risk because airport rides are a good source of income. Waymo undercutting them will reduce the amount of passengers available for pick up. Not saying it’s a bad or good thing. Just that it has real world impact on people and the economy.
Usually you'd have to take the BART one stop then the waymo, which seems to be a common tourist attraction for fresh deplaners. Perhaps the airport was afraid without that step of friction, too many people would try this and cause a waymo-jam
Isn't it by far and wide the most common use of taxi services? It certainly is basically the only time I ever use one.
Waymo getting into that space seems like a pretty big step up in market penetration.
Hope you like traffic!
Waymo can deliver as many rides as Uber with a small fraction of the number of vehicles.
And they're all electric.
How would this reduce or increase traffic? The demand is staying the same.
Presumably the increased supply of "drivers" going to SFO will lower rideshare prices for everyone and make public transit less appealing
Public transit connections to SFO are shit-tier. Rideshare prices have little to do with it.
You can BART from there to downtown, that beats a lot of cities. (yes, yes, BART quality can vary, but still.)
What about the rest of the Bay?
surge pricing FTW!
Can't wait to argue over outsourced call centers to a non human while being stuck in traffic in a dangerous situation
I see a monopoly about to take shape. DOJ/FTC is sleeping on breakup schemes. USDOT should start government/private ventures in this space.
What kind of monopoly?
They are cornering rideshare and automatic vehicles. They are the sole provider of an automatic vehicle.
Fair enough, I suppose they have a monopoly. Good thing that isn't illegal if they're not anti-competitive about it.
They aren't the only autonomous vehicle for hire service. Zoox is operating in Vegas.
Even if they were the only one, it would be odd to classify autonomous rideshare as a distinct market given they compete directly with other vehicle for hire services where they have nothing close to monopoly-like power.
Nobody needs this. This is, in all likelihood, just private enterprise developing technologies for automated warfare.
Have you priced this out compared to a regular taxi or Lyft?
It’s waaaay mo’
Inside SF, my experience is that Uber and Lyft are ~10-15% cheaper than Waymo, but that's before tipping. I don't have to tip a robot, so they work out to nearly identical prices.
You don't have to tip an Uber or Lyft driver either.
Preach.
You also don't have to tip a taxi driver. They get paid for giving rides, it's not an extra service.
Yup, but it is kinda culturally expected to tip. You're right, you dont need to, but then again we do a lot of things just because it is .. polite.
Waymo food delivery will be incredible. You won't even get your food if you don't tip your ubereats.
The inherent problem there is the edges, most food delivery isn't the trip, it's the person getting out of the vehicle and putting it on your doorstep or going through the building. Zipline and their droneports for buildings seem to have the better solution, at least until waymo has some sort of legged robot that can bring the bag the last meter(s)
I think the frustration with tips is so prevalent that the advertising could just be "Skip the tip, simply walk to the street to pick up your order!"
Would work great in suburbs where a robot car could pull in front of home for a minute or two, your food will be bid to another customer if you don't pick it up in 5 minutes. maybe the little robots in NYC are better.
I would argue that the sidewalk robots are too hard to coordinate and not strong enough to hold up against crime, the solution is somewhat closer to my other comment below, a vehicle with maybe 4 or 8 food cells that can fill up at various locations then make its journey around the city. At that point the problem would be idle timeouts and how to handle disgruntled consumers that lost their window for pickup
Aren't the "first meters" also pretty problematic? Are Waymos going to double park in front of a restaurant waiting for someone to come out and put the right order in the right vehicle?
That's easier to do with training, and the business is usually more willing than a consumer as it increases their business. Anecdotally, see how many of them (at least locally to myself) have adopted the doordash/grubhub tablets in their kitchen ordering system. I imagine it would be a co-packing situation with lockers on wheels similar to the vehicle KFC uses in China: https://www.mashed.com/284555/the-futuristic-way-kfc-is-sell...
Uber's NURO seems to be developing a vehicle with a similar form factor as seen on this page: https://www.nuro.ai/first-responders
EDIT: see comment below, uber does not own NURO
Nuro is an independent company from Uber, the latter just has a partnership with and some investment in the former. Uber has similar relationships with more than half the industry at this point.
My anecdotal evidence also has so many incorrect orders that I'm a wee bit less optimistic than you about restaurant-side human handling of the first meters. :)
A lot of restaurants already have dedicated parking spots where they'll bring the food out.
I feel like that's already quite special-case-y.
Certainly wouldn't apply to anywhere near where I currently live (then again, neither does Waymo).
I have a relative in Texas who is looking into leasing a drone to operate for food delivery. Apparently, that's already a thing there? If we could get food/small packages delivered to our building's roof instead of the front door, it would be a huge win for everyone in the building.
Using an otherwise empty 5 person vehicle to move a grocery bag worth of food is pretty stupid though
If it's a robot, why can't it have 10 lockers, and the right one pops open when it arrives to your place?
The longer the route, the harder it is for the food to stay fresh and warm/cold/frozen. It's a trade-off between efficiency, price, and customer satisfaction.
>I don't have to tip a robot
Now that tips are tax free, it's only a matter of time before some clever SV accountants figure out how make everything a tip.
Self-serve ordering terminals already often ask for tips. Presumably to be legal they're being paid to the kitchen staff, but I think sticking to "tips are for workers who have to pretend to like me" is a pretty firm boundary to stick to.
(Also, arbitrarily reclassifying things as tips is hard, because legally 100% of tip revenue has to go to workers, not management, and certainly not the company's investors or coffers).
That's why they're clever accountants.
Tax-free tips paid to robots go to the hardworking AI engineers -> AI engineers voluntarily donate part of their tips to a 501(c)(3) that helps support struggling venture capitalists.
Something like that. We'll work it out the details once the right PAC donations are in place.
Cause what this country needs is to automate away even the gig economy jobs that are out there. Let's keep making a few people rich and screw all the normal people out there.
Why the downvotes? That jobs will be lost is fact. Does this represent an increase in wealth concentration? Obviously. Is that a net bad? I don't know, let's discuss instead of silencing people.
Okay what're the odds on how long it is until there's a stray Waymo on the tarmac. Hopefully with enough warning to divert any planes about to land on it.
Zero. For the same reason there are never any stray civilian cars on the tarmac.
https://www.msn.com/en-us/news/crime/surveillance-video-show...
https://www.cbsnews.com/chicago/news/delivery-driver-secured...
Neither of those are SFO.
Lower than a random human doing that.
Who at Waymo can I speak to about using Waymo’s as an affordable housing solution? I work in commercial real estate and have a handful of affordable projects I am involved in and believe this to be a very interesting solution no one is talking about.
Waymo are toast, Tesla will out scale them in months if not sooner. They can't compete on costs, 100k plus for an ugly Waymo vs < 40k for Tesla model Y or cyber cab.
Waymo makes level 4 autonomous vehicles. Tesla makes level 2 driver assistance systems. Completely different markets.