Waymo has logged 67M miles in the SF Bay Area and 50M miles in LA of autonomous miles driven with a passenger [1]. It looks like the rollout is going quite well in California.
The smallest single-person motorized vehicle is a motorcycle. Leaving stability aside for a minute - one of the issues with electric motorcycles is that the batteries are too small to have range that most riders consider sufficient.
If you have a large enough battery to deliver sufficient range for an electric taxi (driving around most of the day, hopefully, if it is well-utilized), with 4 wheels for stability... I doubt that you could really get significantly smaller where the smaller size would make a big difference.
It does seem necessary to have cameras inside driverless cars but I don't want to if Google is running the show. Your biometrics will probably be stored for "enforcing Google's policies" at a minimum [1].
And then a Waymo dropped teenagers off at the police? I don't care if that was a human, I don't want no snitch car service [2].
Surprisingly low hype for a company running perhaps the most advanced consumer-interactive robots. I’m quite fond of these machines and will periodically rescue them when someone has misused them (door not closed, most commonly). Hope they do well. Very good road participants.
I love Waymo. The only problem I have with it is that it further siphons wages from local economies into Silicon Valley, beyond what Uber/Lyft already did.
I live very close to LAX, and we have hundreds of Waymos in our area. At first i was a little weary, and kinda weirded out by all of them.
They have become completely normal, and cause WAY fewer traffic incidents than human drivers. They are very predictable, and will always let you pull in front of them when changing lanes (unlike a lot of other LA drivers).
I have seen a few incidents where they seem 'stuck', and it was a bit annoying and kind of funny, but again, it does not happen nearly as often as for human drivers.
I also like taking them. It is nice to just get in and relax and not have to talk to anyone or worry about who is going to pick you up. You can control the temperature and music from your phone, you can drop the pin exactly where you want to be dropped off and picked up, and the ride is smooth and comfortable.
We can debate all the knock on effects of Waymos, but the user experience is pretty great.
I am really wondering, how they tackle the long trail problem of unknown situations. Sure they are monitor them constantly but I am asking precisely, how to handle situations that occur ridiculously infrequently: do they sample them 1000x in sim? Or do they use a different model and track uncertainty in the model constantly?
How to get another 0 between the dot and the percent sign in situations where the model makes the wrong choice with a significant probability?
Dear American government regulators: nationwide autonomous driving is proving to be easier than delivering fiber to every home. Why do you hate Americans?
I'm just curious how it works on the legal side: if a human run the red light he gets a fine, who gets fined with a waymo car?
If the car I'm in cause a big damage or kill someone? I guess the damage will be paid by the insurance but whom insurance? Mine or waymos? And what about criminal implications? Insurance doesn't cover thoose for sure.
Elon may have been technically correct about 'if humans can do it with vision only then theres no reason why vision-only cannot work' but turned out to be the wrong strategy for learning how to get there. lidar gives you a constant stream of data to check your vision yourself against and interatively learn. Once trained you turn it off. whereas teslas strategy ended up having to rely on human drivers for error feedback which made the learning loop way longer.
How do the cars do when breaking a rule is necessary? For instance, a car is broken down in the one forward lane of a two-lane road, requiring the waymo to cross the double-yellow line to get around it. Or, a signal is red for so long that obviously its car detector isn’t working.
49 comments
[ 0.23 ms ] story [ 10.1 ms ] threadDoes that include the suburbs?
People hate human drivers more than they hate the idea of AI drivers
https://www.nytimes.com/2026/08/04/business/uber-sexual-assa...
[1] https://waymo.com/safety/impact/
If you have a large enough battery to deliver sufficient range for an electric taxi (driving around most of the day, hopefully, if it is well-utilized), with 4 wheels for stability... I doubt that you could really get significantly smaller where the smaller size would make a big difference.
And then a Waymo dropped teenagers off at the police? I don't care if that was a human, I don't want no snitch car service [2].
[1] https://news.ycombinator.com/item?id=49027861
[2] https://www.404media.co/waymo-called-police-on-teens-san-mat...
They have become completely normal, and cause WAY fewer traffic incidents than human drivers. They are very predictable, and will always let you pull in front of them when changing lanes (unlike a lot of other LA drivers).
I have seen a few incidents where they seem 'stuck', and it was a bit annoying and kind of funny, but again, it does not happen nearly as often as for human drivers.
I also like taking them. It is nice to just get in and relax and not have to talk to anyone or worry about who is going to pick you up. You can control the temperature and music from your phone, you can drop the pin exactly where you want to be dropped off and picked up, and the ride is smooth and comfortable.
We can debate all the knock on effects of Waymos, but the user experience is pretty great.
I prefer not to live like that.
If the car I'm in cause a big damage or kill someone? I guess the damage will be paid by the insurance but whom insurance? Mine or waymos? And what about criminal implications? Insurance doesn't cover thoose for sure.
Jokes aside I always thought this was a crap argument. If you can get more useful signals into a system, it's likely to perform better. Duh!
But never argue with a CEO: their interest is profit, not honest conversations.