Anyone who owns a Tesla with “full self driving” knows how this is going to go. Teslas using only camera vision just don’t have the sensor package and programming to actually perform self driving. I just don’t see this going well for Tesla as it’s more likely to reveal the weaknesses in their technology than be a showcase.
Honestly, I have been hearing this since years now. I was suspicious about it for a long time. But they have been allowing vision based FSD for a long time now and the data shows it's getting better, fast. I am ready to park my skepticism (sometimes cynicism) for some time and give them a benefit of doubt out of optimism. Tesla knows that if things go wrong with Robotaxi launch, they are screwed. They wont' take that risk unnecessarily.
If I was in a car with a human driving that badly, I'd leave ASAP! No-one beyond a learner would traverse in such an uncertain, jerky way. And this is on a day with perfect visibility and no adverse conditions that are visible in the video.
Waymo has had many such incidents like this, but they're still (correctly) considered much safer than human drivers.
Whilst a wheel wobble & veering somewhere it's not supposed to go looks bad, it's very difficult to do worse than the average human driver in terms of safety.
It's actually the opposite. Humans, for all their faults, are amazingly safe drivers. Depending on exactly how you choose to measure things, we achieve anywhere from 5-8 9s of reliability or more.
It's statistically unlikely that we'd see an issue like this on the first day of a limited deployment if FSD was hitting those numbers.
That's true. As part of this deployment they haven't driven enough miles to prove safety, but from their FSD data overall they should have a pretty good understanding of where they stand.
License should be just revoked instantly for the sake of Tesla. At least with not fully autonomous I can override it. They need to get a feedback that they are not ready.
I believe that driving is a privilege, not a right.
However, I'm not saying that their licence should be taken away (we've all made mistakes - I should have made that clear), but if that is representative, it is dangerous. I rode motorbikes for many years, and you can spot people who are dangerous very quickly from their 'car body language'. And an error like that in another circumstance could kill someone.
I knew people whose driving had deteriorated like this (my late mother and her friends spring to mind). They refused to accept they were not capable any more, and she used to do things that terrified me. I had discussions (kindly!) about her driving standards and errors and she refused to accept that she was making the errors she was. And they were not as bad as what I saw in the video.
If I said I've never had a driver's licence, can we get back to the topic at hand?
What's the thesis here, humans are imperfect drivers and therefore we should accept self driving cars driving into incoming traffic, and if anyone objects, focus on their faults instead?
Humans on the aggregate can definitely get better, that's why Norway has 2.14 / 100k traffic-related deaths per year and the United States has 12.84 / 100k.
Norwegians drive half as much. Even if we ignore the other components of the gap, closing that 50% will require rebuilding 300000 square km of urban sprawl into dense urban centers with good public transit. It's a nice thought, but my money is on the robots.
What good is the safety driver doing in the passenger seat? Even the die hard Tesla FSD fans will admit that it isn't uncommon to have a serious disengagement. I give this less than a month before they pull the plug.
In the first instance they're asking to see the passenger's app to verify identity. Seems like a necessary precaution considering what happened to the Waymo vehicles recently.
Presumably if there were an incident they would be trying to remedy any situation.
I really feel we are headed towards a future where humans driving cars will be illegal outside of few exceptions. There will be no traffic as all the robocars will coordinate unselfishly. Car ownership won't be a thing, rides will be a service. Parking spots won't be needed. It's just getting over a safety threshold we are comfortable with.
I have predicted this for years. I think they will make city cores self-driving cars only, with no other cars, or even bicycles because they are too chaotic. China should have done this with those ghost cities, so that they could have dominated with full self driving before other countries.
Someone will own the platform that coordinate car movement, and then all cars will need to pay to get onto this coordination platform that will tell each car how to drive, which route to take, etc. Each car know what all the other cars in its area will be doing, so that mass coordination is possible. This is how you can get a completely full highway but all traveling at 65 mph 1 feet away from each other.
I surely hope not. I'd rather have a city centre without cars than one filled with constantly self-driving cars. Pedestrians and animals are still "chaotic elements". At 105 km/h (still slow by European standards), what happens when a deer suddenly runs onto the motorway with a distance of ~30 cm between the vehicles? Talk about carnage.
I suppose the solution is to seal in all roads then with high walls; what dystopian future.
Imagine all the land being used for parking that wouldn't be needed anymore. Car accidents will be like plane accidents where a full investigation will be launched to improve the car and coordination software.
Yeah, will probably be the same as with horses. Quiet a few of them standing around at the outskirts of the my town. Maybe we'll get an equivalent to horse girls
Car ownership definitely isn’t going away. Sharing a vehicle with the general public gets old fast if you have the means to avoid it.
There are also many reasons people use cars for more than going from point A to point B on a purely transactional basis. Many professions need to leave things in the car or truck like tools or even your laptop. Having to take everything you own into every building in case the self-driving car gets called back home for service or whatever isn’t going to work.
That's probably a thought experiment in business models. Like a franchise. Waymo is experimenting with Uber as it's retailer, essentially. Personally, I'd prefer to do business with Waymo, directly.
If the cost of the Waymo Driver hardware falls to the point where it's not prohibitive for the low duty cycle of a private vehicle, I could see that eventually happening.
I think you're partially right... it'll start with insurance companies, they'll charge people that insist on driving it themselves more for the right (due to safer standards (not that I'm saying this is the case, just that THEY will say it is)).
Then local Gov will charge more to allow non robot cars on the road ( less wear and tear on the roads, fewer roads needed due to more reliably predictable drivers and fewer accidents).
Then lastly manufactures will get to a point that they need to simplify their production range and will pretty much only produce self driving.
lastly, culturally the demand will change. the ipad generation dont want to have learn to drive or have to spend their screen time driving. the damand from them alone will push for self driving cars.
The shoddy past 10-15 years of self driving improvements has led me to believe that this "inevitable future" is at least 20-30 years away, if not longer given that a huge portion of existing cars would need to be effectively outlawed for that that to happen, people and businesses will basically always have legitimate reasons for wanting a human in the loop (and not remotely), and the tech has been worked on actively for decades and still doesn't work well enough to avoid loss of life (which, similar to a terrorist attack, might not be as devastating in numbers as the current state of affairs with humans driving - but will absolutely cause a deeper psychological impact on the general public as there is something seemingly more cruel and dystopian about a company killing people via "error" or cost-savings than a person killing someone by accident).
The psychological distinction is definitely something that was perhaps overlooked. we need to put the blame on some one, for closer. Loosing to some rounding error definitely is haunting
It's already a reality in many places around the world. Most of people don't own a car, just participate in the ride sharing economy. In many places they still have a safety driver like Tesla but there are plenty of places with %100 autonomous driving.
Also authorities are getting giddy when a human tries to drive on railways, so it's effectively illegal to drive in certain places where the ride sharing is the default mode of transportation. It's also very privacy focused, even though there are cameras everywhere you can just buy an anonymous travel pass that you top-up every once in a while. It also allows you to hop between rides for free or at discount.
In the larger cities they often use hyperloop, so you never get stuck in the traffic.
You might not have heard of the biggest one. Apollo Go in China is actually slightly ahead of Waymo in terms of vehicles and miles.
Assuming the Austin robotaxi service continues to go smoothly, I expect Tesla will leapfrog both of them in the next 12 months. Tesla's cost-effective approach (cameras) combined with the fact that they have already scaled vehicle production positions them really well.
Media (and therefore people who trust it too much) will point out places where each of the services goes wrong, but the reality is that likely all of them are already much safer than human drivers if you define "safety" in terms of severe accidents per million miles.
More than a million people die each year from automobile accidents with humans behind the wheel.
> the reality is that likely all of them are already much safer than human drivers if you define "safety" in terms of severe accidents per million miles.
There's no chance this includes Tesla with their disengagements (equivalent to the driver passing out) and even then it's only true under the restricted set of conditions these systems actually operate at compared to human drivers.
Yes but what I was getting at is if the autonomous system is allowed to disengage every time it encounters a "difficult" situation (as Tesla "FSD" does) then its safety record can't be compared to human drivers even in otherwise comparable road conditions.
Human drivers don't get the luxury of disengaging and having a more skilled driver take over when they're struggling. If Tesla FSD drives for 100km before overwhelming glare causes the system to disengage, that'll go on record as "100km driven without accident", but when the human driver is blinded by the same glare and ends up in an accident 5km later, that'll go on record as 1 accident per 5km driven for the human driver.
> they just stop if they ever encounter a situation where they don’t know what to do.
Oh my god that’s terrifying if true. I can think of many situations when driving when slamming on the breaks is the absolute wrong choice. Tesla is pushing this out way before it’s safe enough to operate in public.
Their decision to rely on vision got them billions of miles of training data from every corner of the USA. It did not get them point clouds, but monocular depth estimation works extremely well these days, so would it really have been that valuable? It looks to me like they made a shrewd bet and won big.
Seems like a mistake not to be a sensor maximalist. Why not try and get as many data streams onboard as possible. Seems like the generalized robot OS would be better if it was a data fusion platform.
Calling this 'Robotaxi' rides seems somewhat disingenuous given that they are Model Ys with a full seating provision. People will be conflating this with the (IMO a very bad choice) 2-seat design seen on the demo months ago with 2 doors and no provision for more than two people (which wouldn't work too well if the drives are being chaperoned, reducing seating to one).
This conflation and mixing of words is something which Tesla seems to do intentionally (Full Self Driving, Autonomy, etc), or am I being cynical about that?
I agree that 'cybercab' and 'robotaxi' are similar names, but what ulterior motive would it serve to intentionally confuse people about the difference? Seems like it's just bad naming. Cybercab isn't late or anything, it was announced for 2026 at its unveiling. And Robotaxi is an accurate and descriptive name for a self-driving Model Y that you can hail.
Difference between this and Cruise or Uber is that Elon is in charge of regulation. A driverless Tesla could plow through a farmer's market and service wouldn't even stop for the afternoon.
Let’s give it a little bit of time to see, but gating the release of self driving taxis to sycophants is not the choice you make when you are confident with a product…
Are you looking at the feeds from the cars to reach 32? Each screen is from the cameras around a single car, so you're actually only seeing 5 cars from that.
If you can eventually guarantee you’ll get a robotaxi instead of a car with a driver when you call one (unlike Waymo in Austin, now, thanks to their dumb partnership with Uber), they’ll likely do ok in the longer term.
I'm rooting for Tesla being able to provide some driverless taxi competition to Waymo. Hopefully both can increase supply enough to allow for a guaranteed robotaxi booking, although anecdotally I haven't found it that hard to get a Waymo through the Uber app depending on the time.
I’ve only tried twice (after using it ~30 times during the preview directly through the Waymo app) - neither time resulted in a robotaxi though, and one resulted in a driver that smelled so bad I’ve sworn off ride sharing until the end of summer.
Hopefully the Tesla app forces Waymo to reconsider their deal with Uber altogether and just run it themselves like they do in San Francisco.
"Tesla has self-driven 4B miles in every corner of the USA, Waymo has only self-driven 0.1B miles in select cities, how pathetic! They are so far behind Tesla it isn't even funny!"
- Me, if I wanted to be equally ham-handed in throwing the comparison for Tesla.
In reality, we are seeing two bets on two different approaches. Do you scale up supervised driving to maximize data collection/diversity and then go unsupervised? Or do you go unsupervised and then scale up problem solving as you go? The cool thing is that both of these approaches are being tried so we will find out. If you want to place a bet of your own, you can find the casino in your favorite brokerage app!
In any case, the videos of what is possible with supervised FSD are quite amazing, certainly not "pathetic," and what remains to be seen is if they can successfully navigate the supervised->unsupervised jump, which is certainly not trivial.
That's not usually the argument. It's that LiDar has historically been working more reliably than tech that doesn't use it - and reliability is pretty important in that industry.
In any case I'm curious what gives you such blind faith that Tesla's vision-only tech will work out. There's a lot of really practical and frankly obvious reasons why it hasn't been working for a very long time and similarly lot's of videos (which are aggressively taken down by Tesla or can't even be posted due to shady customer NDA's) of their self-driving equipment getting dangerously close to murdering people. And of course, there are the _actual_ cases of Tesla's self-driving tech just you know, straight up murdering people.
You can shift blame to the humans not being attentive enough to take control when self-driving mistakes come up, but then you've lost the broader argument that self-driving Tesla's are "ready", even if you're right about the former.
Edit because fuck-you-HN-comment-limit-soft-ban:
(response to child comment below):
> Obviously, only vision and hearing are required for driving
Human vision is drastically better than even the most advanced computer vision systems we have or are likely to have any time soon - especially when there are realtime and hardware constraints. That's before you introduce the relatively (!) shoddy approximations that even SOTA vision models learn via SGD.
> I'm curious what gives you such blind faith that Tesla's vision-only tech will work out.
Obviously, only vision and hearing are required for driving. The question isn't "can it work," the question is "which approach will be economically viable on what time scale, and is there a winner-takes-all dynamic that excludes the approach that comes second."
I've never understood the argument "humans only use vision and hearing so self driving cars should too". We don't constrain what we do with machines to what humans do.
Humans don't have wheels either. Is Tesla planning to replace wheels in their cars with legs, Flintstones style?
One example from the first day, where the vehicle spends a bit of time driving into oncoming traffic:
https://youtu.be/_s-h0YXtF0c?t=0h7m15s
Anyone who owns a Tesla with “full self driving” knows how this is going to go. Teslas using only camera vision just don’t have the sensor package and programming to actually perform self driving. I just don’t see this going well for Tesla as it’s more likely to reveal the weaknesses in their technology than be a showcase.
Given that and given that they know the data better than anyone, what's your hypothesis as to why they would deploy if the data isn't in their favor?
Musk needs to keep hyping the next big thing to justify the insane valuation. So it was going to launch regardless of how ready or safe it was.
Still seems nuts to me, they're pretending Waymo doesn't already regularly drive without supervision.
https://en.wikipedia.org/wiki/List_of_predictions_for_autono...
Honestly, I have been hearing this since years now. I was suspicious about it for a long time. But they have been allowing vision based FSD for a long time now and the data shows it's getting better, fast. I am ready to park my skepticism (sometimes cynicism) for some time and give them a benefit of doubt out of optimism. Tesla knows that if things go wrong with Robotaxi launch, they are screwed. They wont' take that risk unnecessarily.
If I was in a car with a human driving that badly, I'd leave ASAP! No-one beyond a learner would traverse in such an uncertain, jerky way. And this is on a day with perfect visibility and no adverse conditions that are visible in the video.
Waymo has had many such incidents like this, but they're still (correctly) considered much safer than human drivers.
Whilst a wheel wobble & veering somewhere it's not supposed to go looks bad, it's very difficult to do worse than the average human driver in terms of safety.
It's actually the opposite. Humans, for all their faults, are amazingly safe drivers. Depending on exactly how you choose to measure things, we achieve anywhere from 5-8 9s of reliability or more.
It's statistically unlikely that we'd see an issue like this on the first day of a limited deployment if FSD was hitting those numbers.
That's true. As part of this deployment they haven't driven enough miles to prove safety, but from their FSD data overall they should have a pretty good understanding of where they stand.
License should be just revoked instantly for the sake of Tesla. At least with not fully autonomous I can override it. They need to get a feedback that they are not ready.
[dead]
[flagged]
They shouldn't be on the road either. That's not a positive argument for the Tesla experience, it's a negative one for whoever was driving that car.
I don't like what that person did, but I don't think their license should be taken away for it.
I believe that driving is a privilege, not a right.
However, I'm not saying that their licence should be taken away (we've all made mistakes - I should have made that clear), but if that is representative, it is dangerous. I rode motorbikes for many years, and you can spot people who are dangerous very quickly from their 'car body language'. And an error like that in another circumstance could kill someone.
I knew people whose driving had deteriorated like this (my late mother and her friends spring to mind). They refused to accept they were not capable any more, and she used to do things that terrified me. I had discussions (kindly!) about her driving standards and errors and she refused to accept that she was making the errors she was. And they were not as bad as what I saw in the video.
And so it’s ok to break traffic rules? In my country, as a human this will get you easily a one month license suspension.
No, just a bit of perspective. I'm sure the video record of your car would be squeaky clean with no violations whatsoever.
If I said I've never had a driver's licence, can we get back to the topic at hand?
What's the thesis here, humans are imperfect drivers and therefore we should accept self driving cars driving into incoming traffic, and if anyone objects, focus on their faults instead?
We allow imperfect humans, we should allow similarly imperfect robots. The robots will get better. The humans, in aggregate, will not.
Humans on the aggregate can definitely get better, that's why Norway has 2.14 / 100k traffic-related deaths per year and the United States has 12.84 / 100k.
Norwegians drive half as much. Even if we ignore the other components of the gap, closing that 50% will require rebuilding 300000 square km of urban sprawl into dense urban centers with good public transit. It's a nice thought, but my money is on the robots.
Robots are not going to fix other problems with sprawl. People there are focusing on wrong solutions.
What good is the safety driver doing in the passenger seat? Even the die hard Tesla FSD fans will admit that it isn't uncommon to have a serious disengagement. I give this less than a month before they pull the plug.
> What good is the safety driver doing in the passenger seat?
The same as a driving instructor. He has a brake pedal, and he can pick up the steering wheel if necessary.
In the first instance they're asking to see the passenger's app to verify identity. Seems like a necessary precaution considering what happened to the Waymo vehicles recently.
Presumably if there were an incident they would be trying to remedy any situation.
I really feel we are headed towards a future where humans driving cars will be illegal outside of few exceptions. There will be no traffic as all the robocars will coordinate unselfishly. Car ownership won't be a thing, rides will be a service. Parking spots won't be needed. It's just getting over a safety threshold we are comfortable with.
I have predicted this for years. I think they will make city cores self-driving cars only, with no other cars, or even bicycles because they are too chaotic. China should have done this with those ghost cities, so that they could have dominated with full self driving before other countries.
Someone will own the platform that coordinate car movement, and then all cars will need to pay to get onto this coordination platform that will tell each car how to drive, which route to take, etc. Each car know what all the other cars in its area will be doing, so that mass coordination is possible. This is how you can get a completely full highway but all traveling at 65 mph 1 feet away from each other.
No bicycles? That is ridicules. Many city’s are now trending towards people centric design. Bikes are a key part of that
I surely hope not. I'd rather have a city centre without cars than one filled with constantly self-driving cars. Pedestrians and animals are still "chaotic elements". At 105 km/h (still slow by European standards), what happens when a deer suddenly runs onto the motorway with a distance of ~30 cm between the vehicles? Talk about carnage.
I suppose the solution is to seal in all roads then with high walls; what dystopian future.
Imagine all the land being used for parking that wouldn't be needed anymore. Car accidents will be like plane accidents where a full investigation will be launched to improve the car and coordination software.
A lot of people really like driving and are pretty into cars. Maybe in 100+ years but I think this is sci-fi in the near term.
Yeah, will probably be the same as with horses. Quiet a few of them standing around at the outskirts of the my town. Maybe we'll get an equivalent to horse girls
> Car ownership won't be a thing
Car ownership definitely isn’t going away. Sharing a vehicle with the general public gets old fast if you have the means to avoid it.
There are also many reasons people use cars for more than going from point A to point B on a purely transactional basis. Many professions need to leave things in the car or truck like tools or even your laptop. Having to take everything you own into every building in case the self-driving car gets called back home for service or whatever isn’t going to work.
Google is exploring bringing the Waymo driver to privately owned cars as well.
That's probably a thought experiment in business models. Like a franchise. Waymo is experimenting with Uber as it's retailer, essentially. Personally, I'd prefer to do business with Waymo, directly.
If the cost of the Waymo Driver hardware falls to the point where it's not prohibitive for the low duty cycle of a private vehicle, I could see that eventually happening.
I think you're partially right... it'll start with insurance companies, they'll charge people that insist on driving it themselves more for the right (due to safer standards (not that I'm saying this is the case, just that THEY will say it is)).
Then local Gov will charge more to allow non robot cars on the road ( less wear and tear on the roads, fewer roads needed due to more reliably predictable drivers and fewer accidents).
Then lastly manufactures will get to a point that they need to simplify their production range and will pretty much only produce self driving.
lastly, culturally the demand will change. the ipad generation dont want to have learn to drive or have to spend their screen time driving. the damand from them alone will push for self driving cars.
The shoddy past 10-15 years of self driving improvements has led me to believe that this "inevitable future" is at least 20-30 years away, if not longer given that a huge portion of existing cars would need to be effectively outlawed for that that to happen, people and businesses will basically always have legitimate reasons for wanting a human in the loop (and not remotely), and the tech has been worked on actively for decades and still doesn't work well enough to avoid loss of life (which, similar to a terrorist attack, might not be as devastating in numbers as the current state of affairs with humans driving - but will absolutely cause a deeper psychological impact on the general public as there is something seemingly more cruel and dystopian about a company killing people via "error" or cost-savings than a person killing someone by accident).
The psychological distinction is definitely something that was perhaps overlooked. we need to put the blame on some one, for closer. Loosing to some rounding error definitely is haunting
It's already a reality in many places around the world. Most of people don't own a car, just participate in the ride sharing economy. In many places they still have a safety driver like Tesla but there are plenty of places with %100 autonomous driving.
Also authorities are getting giddy when a human tries to drive on railways, so it's effectively illegal to drive in certain places where the ride sharing is the default mode of transportation. It's also very privacy focused, even though there are cameras everywhere you can just buy an anonymous travel pass that you top-up every once in a while. It also allows you to hop between rides for free or at discount.
In the larger cities they often use hyperloop, so you never get stuck in the traffic.
I honestly thought Tesla would beat everyone to the punch here but it seems like they just stalled out what - five or so years ago?
Waymo seems to be the furthest ahead in my complete outsider opinion. I feel like everyone else is a way distant second.
You might not have heard of the biggest one. Apollo Go in China is actually slightly ahead of Waymo in terms of vehicles and miles.
Assuming the Austin robotaxi service continues to go smoothly, I expect Tesla will leapfrog both of them in the next 12 months. Tesla's cost-effective approach (cameras) combined with the fact that they have already scaled vehicle production positions them really well.
Media (and therefore people who trust it too much) will point out places where each of the services goes wrong, but the reality is that likely all of them are already much safer than human drivers if you define "safety" in terms of severe accidents per million miles.
More than a million people die each year from automobile accidents with humans behind the wheel.
> the reality is that likely all of them are already much safer than human drivers if you define "safety" in terms of severe accidents per million miles.
There's no chance this includes Tesla with their disengagements (equivalent to the driver passing out) and even then it's only true under the restricted set of conditions these systems actually operate at compared to human drivers.
It wouldn't make sense for the robotaxis to ever disengage in the same way that you might've seen in FSD videos.
I'm guessing that they just stop if they ever encounter a situation where they don't know what to do.
Yes but what I was getting at is if the autonomous system is allowed to disengage every time it encounters a "difficult" situation (as Tesla "FSD" does) then its safety record can't be compared to human drivers even in otherwise comparable road conditions.
Human drivers don't get the luxury of disengaging and having a more skilled driver take over when they're struggling. If Tesla FSD drives for 100km before overwhelming glare causes the system to disengage, that'll go on record as "100km driven without accident", but when the human driver is blinded by the same glare and ends up in an accident 5km later, that'll go on record as 1 accident per 5km driven for the human driver.
> they just stop if they ever encounter a situation where they don’t know what to do.
Oh my god that’s terrifying if true. I can think of many situations when driving when slamming on the breaks is the absolute wrong choice. Tesla is pushing this out way before it’s safe enough to operate in public.
Yeah, but Tesla is headed by a ketamine-addicted imbecile.
Will sound engineering prevail over brain rot in the C-suite? I am skeptical.
I hope he gets shut down so hard he winds up living in a box in an alley. He's a billionaire Neo-Nazi.
Their decision to entirely rely on computer vision seems unwise
"LiDAR is gay"
- Elon, probably
Their decision to rely on vision got them billions of miles of training data from every corner of the USA. It did not get them point clouds, but monocular depth estimation works extremely well these days, so would it really have been that valuable? It looks to me like they made a shrewd bet and won big.
Short term, perhaps, long term they might end up with a more generalized robot operating system. It’s definitely more flexible
Seems like a mistake not to be a sensor maximalist. Why not try and get as many data streams onboard as possible. Seems like the generalized robot OS would be better if it was a data fusion platform.
It's hard to attract and retain key talent when the CEO is going around acting all Nazi-curious. Good people who can leave are likely to do so.
That may be true, but Tesla stalled out a long time before Musk's rightward turn.
Nazi-curious is just a state of substance-curious in this case IMHO
Oh well in that case it's totally excusable /s
Oh no you misunderstood. He’s been insufferable for many years before this.
Calling this 'Robotaxi' rides seems somewhat disingenuous given that they are Model Ys with a full seating provision. People will be conflating this with the (IMO a very bad choice) 2-seat design seen on the demo months ago with 2 doors and no provision for more than two people (which wouldn't work too well if the drives are being chaperoned, reducing seating to one).
This conflation and mixing of words is something which Tesla seems to do intentionally (Full Self Driving, Autonomy, etc), or am I being cynical about that?
I agree that 'cybercab' and 'robotaxi' are similar names, but what ulterior motive would it serve to intentionally confuse people about the difference? Seems like it's just bad naming. Cybercab isn't late or anything, it was announced for 2026 at its unveiling. And Robotaxi is an accurate and descriptive name for a self-driving Model Y that you can hail.
One accident and the knives are gonna be out. But if it works the world will change
Difference between this and Cruise or Uber is that Elon is in charge of regulation. A driverless Tesla could plow through a farmer's market and service wouldn't even stop for the afternoon.
In summary:
> 10-or-so cars
> human driver behind the wheel (except in this beta)
> invitation based (apparently a very limited audience)
> geofenced not only to a city, but to a small handful of neighborhoods
> early reports suggest disengagement requires remote re-engagement
I hope they get there, more competition in this space is good. But, this is pathetic. They're so far behind Waymo it isn't even funny.
Let’s give it a little bit of time to see, but gating the release of self driving taxis to sycophants is not the choice you make when you are confident with a product…
I understand the thinking here, but this is too early to be too negative, IMO. For the first day, I think this looks like a reasonable first step.
If they are still doing this in 3 months, it'd be a bad sign, of course. Their plan is for rapid growth next year.
We'll see if they are able to do that.
"next year" is Musk speak for "not going to happen anytime soon"
Sounds like there’s actually around 32ish, based on photos from the “war room” where all the engineers have live feed from all the cars and 112 rides ? https://x.com/aelluswamy/status/1936865682810946035
They had said 10 - 12 prior to launch.
Are you looking at the feeds from the cars to reach 32? Each screen is from the cameras around a single car, so you're actually only seeing 5 cars from that.
No human behind the wheel but there are still people in the passenger seat
The beta today was without a driver behind the wheel. Granted, the geofence is much smaller than Waymo's in Austin.
If you can eventually guarantee you’ll get a robotaxi instead of a car with a driver when you call one (unlike Waymo in Austin, now, thanks to their dumb partnership with Uber), they’ll likely do ok in the longer term.
I'm rooting for Tesla being able to provide some driverless taxi competition to Waymo. Hopefully both can increase supply enough to allow for a guaranteed robotaxi booking, although anecdotally I haven't found it that hard to get a Waymo through the Uber app depending on the time.
I’ve only tried twice (after using it ~30 times during the preview directly through the Waymo app) - neither time resulted in a robotaxi though, and one resulted in a driver that smelled so bad I’ve sworn off ride sharing until the end of summer.
Hopefully the Tesla app forces Waymo to reconsider their deal with Uber altogether and just run it themselves like they do in San Francisco.
It's only day 1.
Day 1 or year 10. Depends on perspective.
Development and releases are two different things.
"Tesla has self-driven 4B miles in every corner of the USA, Waymo has only self-driven 0.1B miles in select cities, how pathetic! They are so far behind Tesla it isn't even funny!"
- Me, if I wanted to be equally ham-handed in throwing the comparison for Tesla.
In reality, we are seeing two bets on two different approaches. Do you scale up supervised driving to maximize data collection/diversity and then go unsupervised? Or do you go unsupervised and then scale up problem solving as you go? The cool thing is that both of these approaches are being tried so we will find out. If you want to place a bet of your own, you can find the casino in your favorite brokerage app!
In any case, the videos of what is possible with supervised FSD are quite amazing, certainly not "pathetic," and what remains to be seen is if they can successfully navigate the supervised->unsupervised jump, which is certainly not trivial.
Arc de Triomphe: https://youtu.be/o2xKpbKZLVA?t=7
Busy China: https://youtu.be/ybBpRN4Hqbc?t=13
Manhattan: https://www.youtube.com/watch?v=qafr3RrJRfU
[flagged]
That's not usually the argument. It's that LiDar has historically been working more reliably than tech that doesn't use it - and reliability is pretty important in that industry.
In any case I'm curious what gives you such blind faith that Tesla's vision-only tech will work out. There's a lot of really practical and frankly obvious reasons why it hasn't been working for a very long time and similarly lot's of videos (which are aggressively taken down by Tesla or can't even be posted due to shady customer NDA's) of their self-driving equipment getting dangerously close to murdering people. And of course, there are the _actual_ cases of Tesla's self-driving tech just you know, straight up murdering people.
You can shift blame to the humans not being attentive enough to take control when self-driving mistakes come up, but then you've lost the broader argument that self-driving Tesla's are "ready", even if you're right about the former.
Edit because fuck-you-HN-comment-limit-soft-ban:
(response to child comment below):
> Obviously, only vision and hearing are required for driving
Human vision is drastically better than even the most advanced computer vision systems we have or are likely to have any time soon - especially when there are realtime and hardware constraints. That's before you introduce the relatively (!) shoddy approximations that even SOTA vision models learn via SGD.
> I'm curious what gives you such blind faith that Tesla's vision-only tech will work out.
Obviously, only vision and hearing are required for driving. The question isn't "can it work," the question is "which approach will be economically viable on what time scale, and is there a winner-takes-all dynamic that excludes the approach that comes second."
I've never understood the argument "humans only use vision and hearing so self driving cars should too". We don't constrain what we do with machines to what humans do.
Humans don't have wheels either. Is Tesla planning to replace wheels in their cars with legs, Flintstones style?
WayToLittle, wayToLate, wayMore already on the road in Cal!