They're growing fast, but I do wonder what they see as the bottlenecks: municipal approval, training, manufacturing? It's really unfortunate that lobbying is slowing self-driving rollout in some of the biggest and most important cities: NYC, Chicago, DC.
So weird little things for scaling, but overall it was a good ride till that point. Also, the dropoff was at a hospital/clinic and Lyft for the return trip had no issue with going into the property to the actual destination.
The app has changed the method a few times. It used to show green or blue segments for valid spots, but today as it is, I just pick up the pin, drop it, hope for the best. It also demonstrates which way the car will be facing.
Dropping a precise pin on the map is no guarantee that it will actually stop right there, depending on the situation on the ground when it arrives. But, you can still adjust the pin while the car waits, or you can hit the "Pull Forward" button to try and improve it.
The app has also been updated to disallow adjustment of multiple dropoffs beyond the initial stop. So if you want a precise pin-drop in a multi-stop trip, you need to intervene along the way, instead of during your initial setup.
There was a recent case reported in the news, where people were trying to Waymo to the San Bruno airport, but their pin got dropped off in a residential neighborhood. This was apparently because Waymo doesn't directly serve the airport's lanes yet, so the pin dropped as close as possible, but the residents of that street were nonplussed about the results. So were the airline passengers who were lugging luggage!
Those are also cities with a great public option for transit (unlike, say, Dallas or the Bay Area where different municipalities squabble). One might laud New York driving people away from car culture when the subways/buses can serve more people, greener.
The current mayor of NYC is a DemSoc so he might have some ideas about keeping NYC's money in the pockets of middle-class NYC citizens.
It is so disappointing to see the tech class advocate for laws that keep jobs that are unhealthy and lead to nowhere (cabbies) instead of moving the country forward (and those people getting other, hopefully better, jobs). But again, that type of short-sighted, zero-sum/socialist thinking is becoming more common, sadly.
No contest.
A typical jaguar Waymo usually carries up to 4 people (usually just one), weighs 2.1 tonnes and is 5.2m long.
So the double decker bus should be more environmentally efficient with maybe 5 passengers making independent journeys, though operational costs for paying a driver probably make the Waymo more financially efficient in that situation.
I suspect robotaxis will become very important in suburban areas with distributed jobs where buses don’t work well. In busy cities there simply isn’t the roadspace to replace buses with robotaxis (or any other form of car) for commuting.
A more reasonable taxi for single riders is going to weigh around 1.5 tons (e.g. the prototype Tesla Robocab).
You also incorrectly assume that the efficiency depends linearly on weight. It doesn't, it depends more on aerodynamics, speed, and the number of stops. London buses (11 tons) average around 2.2 kWh/mi, Tesla Model 3 is at 0.21 kWh/mi while weighing 1.8 tons.
People here only ride the bus when there’s no other option for them.
Waymo could make a bus too anyway, especially after they’ve been in an area a while and have data about where they’d have the most impact. Internet at large would probably decry it as enshittification.
Yes, if you live within about 1 mile of downtown and also work there. If you live downtown and your job is in Beaverton, you're likely in for a 90 minute commute. Or if you live in Beaverton and your office is downtown. Portland's mass transit is adequate for a subset of people and completely impractical for most.
Source: I've lived in and around Portland for 50 years. Worked downtown much of it, and have lots of friends and former colleagues who used public transit. You have to really want to not have a car because the time commitment is real.
Oh hell no. Cars are bad enough from the damage they can cause by kinetic energy, but a bus? Especially a bus where people don't wear seatbelts or, worse, are standing?
But wait, there's so much more!
A single bus/subway ride costs about $20, which is a surprisingly consistent figure for various US cities. In NYC it's $17, for example. I'm talking about the _true_ cost, as in: "yearly operational budget divided by the number of trips". And most of this cost comes from 3-4 drivers that every bus needs, not from the hardware itself.
Price is a pretty reliable proxy for CO2 emissions for general market goods and services. And with these kinds of numbers, personal cars are competitive. Even if we factor in subsidized road maintenance cost ( https://taxfoundation.org/data/all/state/state-infrastructur... ).
If we now add mild carpooling, so that a single robotaxi drives 2-3 people, then classic transit just dies. It's like comparing horse-drawn carriages with steam trains.
Drivers emit a lot of CO2, just like every person in the US. A driverless car allows these emissions to be reallocated to other more productive areas.
I've also seen the autonomous waymo vans, which seem nice too.
Is there a "theoetical" ahead and a "practical" ahead?
How do the finances look and does financial solvency count? will someone run out of money first?
There have been two waymo articles (about sensors, and now about hardware), is this a sign they are worried?
I am actually wondering.
Yes. Waymo is doing something like 400K fully autonomous rides per week and Tesla is doing hundreds or low thousands of supervised rides.
In the last two years there have been dozens of incidents of Waymos obstructing first responders in life-threatening situations. Every major metropolitan area Waymo has deployed in, the police departments have reported extensive issues with Waymos obstructing officers and emergency vehicles, and worse, their remote support staff often being unable to do anything about it.
If that's "so far ahead" of everyone else, then we are pretty fucked.
I mean drafting code on my phone with chatgpt in a Waymo is legitimately some Jestons shit and I’m not hearing otherwise.
Amazing
I don't know if we'll get there eventually, but if we do, I won't be surprised if it's because of Waymo.
I wish more people would experience it and see that the future is now.
It reminds me of putting on noise-cancelling headphones in an already quiet room. You suddenly gain an appreciation for just how much quieter things could've been thanks to the contrast.
Similarly, having a vehicle accelerate/decelerate consistently, always taking turns with the same sharpness, etc.. all of it builds to this subtle feeling of the car ride fading into the background of the experience. You didn't realize those were features of a drive you could appreciate until experienced.
This does seem at least a bit intentional on Waymo’s part. They appear to have adopted an under promise, over deliver mindset (unlike some other robotaxi alternatives that come to mind). They’ve seen a lot of their competitors (Uber, Cruise, etc) capture a lot of ultimately negative attention and folded as a result; I would probably try to avoid the limelight if I were them too. The tech speaks for itself.
Tesla Robotaxis need to hog the media spotlight because they're trying to prop up the stock price, not because they work.
Both self-driving car and general purpose AI robots are emerging technologies right now. My bet is that eventually, the HW will converge accordingly.
It would just be cool to see what these guys have cooked up dev tools wise since Google has always been famed for good developer experience (often to such a degree that Xooglers are useless outside the exosuit).
It would rely on them being nice enough to leave at least some binaries on the thing though. Because I'm not sure how would one even get an in on how to boot the things without.
uhhh i've been in a situation where a remote human took over before
it's rare but it can happen (or at least it could 1.5 years ago-ish)
the specific situation was a tight alleyway with cars parked on either side and people doing things like opening car doors and crossing randomly. the waymo obviously wanted a much wider safety margin than what was available and stopped to wait for things to clear up. the remote human driver instead drove more aggressively towards the pedestrians so they would clear the road.
is it me or their example photo doesn't look even like 1 high-resolution camera (and very noisy at that, looks like a pretty high ISO setting). Their example looks similar and i'd say a bit worse than what my Fuji camera did 25 years ago in the dark after i'd adjust dynamic range of the image in GIMP. (and the cameras on the Waymo cars i see around have about the same lens size as my Fuji)
If it is really a showcase of their visual pipeline than it makes sense that they are still so reliant on lidar - which is pretty expensive because of the number of rays you have to support to maintain good situation-awareness and which is still hard and expensive to scale to get good resolution. That also means that their tech can't really be put onto other platforms like drones for example.
Also, what looks good to us may not be "good" to a machine.
But I'm really just guessing in the dark here.
of course, whatever "bettering" algorithm makes sense, the neural network would probably emerge it inside itself as a result of training. My comment was more about hardware in the sense that "garbage in - garbage out", and so you'd want to have a decent hardware which is available and pretty cheap today while Waymo seems to be using, i'm stating it politely :), very-very-commodity level.
>Also, what looks good to us may not be "good" to a machine.
the kernels of the first layers of visual CNNs (at least the ones i looked at) converge to the Gabor-like kernels what are in the first layers of the human visual cortex. Bunch of years ago, before deep-learning CNN, etc. i saw an article (in neurophysiology field) proving an optimality of such kernels for those visual tasks. Of course things may and probably diverge a lot in the deeper layers, yet deep dreaming, etc. suggests to me that the nature of the divergence may be less severe than expected.
> New York abandoned its proposal to allow self-driving taxis earlier this year. New York City’s new mayor, Zohran Mamdani, has said he would heavily weigh the interests of taxi drivers in deciding on rules for the technology.
0: https://www.nytimes.com/2026/06/17/technology/waymo-driverle...
Do you think they will?
I don’t think there’s any mystery as to why AI is not a good marketing term.
Apple avoids “AI” hype at WWDC keynote by baking ML into products (2023)
https://arstechnica.com/information-technology/2023/06/at-ap...