If they can operate in London, they will have really shown that autonomy is working. London is full of two-way roads that are only one car width wide, roads were you can just get stuck and have to fully back out, vehicles entirely blocking roads requiring complex cooperation between drivers to negotiate, etc. And that's not even considering the complex logic of figuring out where you can stop to let someone out.
> If they can operate in London, they will have really shown that autonomy is working
Go back through Waymo’s historic announcements on HN. This is said every time.
Autonomy works. Waymo has solved it. There isn’t yet a number 2, though China has strong candidates. But where you can find Waymo, it works so well that we need to see it in a familiar context to believe it really exists.
It's unclear as to where in London this is being rolled out, but central London traffic is notoriously slow, which may help. But we also don't have the concept of jaywalking - you can cross wherever, whenever and there is quite the cycling infrastructure to deal with.
The linked article is bad. It would've been better to link to the official announcement or page (https://waymo.com/waymo-in-uk):
"select boroughs in London, including Barking and Dagenham, Brent, Camden, Ealing, Royal Borough of Greenwich, Hackney, Hammersmith and Fulham, Haringey, Islington, Royal Borough of Kensington and Chelsea, Lambeth, Lewisham, Newham, Redbridge, Southwark, Tower Hamlets, Waltham Forest, Wandsworth, City of Westminster, and the City of London."
I mean, it's now technically true (since some point in 2024) in NYC, but historically it was merely practice to not uniformly enforce this rule and that's basically a more extreme version of the reality across the US.
However unlike a cop deciding whether to fine people for "jay walking" the Waymo doesn't care whether you're the mayor's grandmother or a wise-talking black teenager if you're in its path, those are both humans and it's not allowed to hit humans.
There aren't going to be many humans in the street in outer Phoenix because where would they even be going? But the Mission, just like Leicester Square, has plenty of pedestrians who might just run into the street in front of you given any reason or opportunity and you need to be ready for that.
So I agree this isn't novel for Waymo, but while technically jay walking is now legal in NYC that's not why Waymo (which also has pencilled in an NYC launch) needs to care about it already.
Your comment is a slight of hand. Yes it "works" but it is an MVP. Everywhere it's rolled out people say yes it drives, but it drives like a rookie that can follow the rules but doesn't have any of the predictive power that every driver who doesn't suck has some of.
Like for example how traffic will often modulate its spacing or time its lane changes to reduce issues with merging or exiting traffic and certain intersections with most traffic doing the same thing become an efficient repetitive cycle, Waymo doesn't "get it".
So it works, in the same way a newly licensed teenager "works". It's no cabbie.
Edit: It's been over a year since I've ridden in one. Good to hear it's potentially better now.
After taking both Uber and Waymo in San Francisco I think I disagree with this. Only caught a few so maybe I didn't see the bad sides, but from what I saw it was able to predict flows and navigate far more smoothly than humans in many situations. Really comfortable and overall safe feeling ride.
It's different than human drivers for sure, but to me at least it's better.
I agree with MVP part, my understanding is that there's still a lot of Wizard of Oz stuff in terms of regularly mapping and remapping its routes and having a lot of human operators remotely checking and maybe controlling the fleet, but I'm impressed personally.
> Everywhere it's rolled out people say yes it drives, but it drives like a rookie
When did you last update this hypothesis?
The Waymos I’ve been in creep, honk and modulate their aggressiveness quite naturally. In the cities they operate in, they’re a premium product to cabs.
The part where you’re right is on freeways. But my point is that tends to be ignored when folks gatekeep “real” autonomy. Instead, it’s some random peculiarity of their city which humans traverse at low speeds. Exactly the thing Waymo has mastered.
All the existing places Waymo operate are in the US though, as OP says London is very different. Will be interesting to see if they can get it working or not
Sure, but in terms of traffic difficulty they've done like level 1, 2, 3, 4, 5 and now they're jumping to 100! This is different.
It's not the most difficult place to drive (good luck in India, Turkey, southern Italy, etc.), but it's still far more challenging than any American city.
And it needs fundamentally new capabilities like being able to negotiate with other drivers visually and read implicit signals. You can't do it all just by following what the traffic lights say.
I've driven extensively in both, and London's hands-down more challenging°. Boston drivers are significantly more aggressive, though, so Waymo might find the reverse. I dunno.
——
°Which makes it more fun: I love driving in London, but I'm a weirdo. My biggest regret about my last / first visit to Delhi is that I let my wife talk me out of driving. It may indeed have been a prudent choice, but I think I was there long enough to "get" enough to manage next time.
Residents of every city claim that they have the craziest drivers or toughest streets to navigate in. London isn't really that materially harder to drive in than San Francisco.
But I've lived in both places and London is very different than SF. I'd say the UK has better drivers on average (and much more strict licensing requirements), but driving in London is much more challenging due to the tiny roads you have to navigate. There is no road in SF that is as hard to navigate as the average suburban London two-way traffic single car width road with parking on both sides.
An I'm not saying London is "the worst" by any means. It's nothing like driving in Vietnam or India. But it is very different to SF.
Driving in London is kind of weird these days in that you feel almost stationary. It's typically wait at lights for 2 min, drive at 15-20 mph for 40 seconds, repeat. I've mostly given up on it and use an ebike instead.
That is just nonsense, sorry. San Francisco is a modern layout (grid) and London is an ancient city with road networks based on thoroughfares that are 500+ years old.
It may be true that all major cities have their own quirks, but London has significantly more complexity than San Francisco or any US grid based city with super wide roads.
Also, the US bought into the ‘Car is King’ idea whereas that’s never really been the case in the UK outside of a few places like Birmingham. It’s generally harder to be a driver in the UK.
Whether that causes significant problems for Waymo, who knows? But I am also of the opinion that if it works in London then that’s a pretty powerful tell that they’ve got it right. We’ll, at least for places where drivers generally stick to the rules.
How do autonomous driverless cars handle the stand off that occurs when two cars meet on a narrow London street and both parties are of the opinion that the other has to back up 50 meters to let them through? Will the car instruct the passenger to get out and do the arm waving and swearing required for the right of way negotiation?
Robo-taxis can decide not to travel through these roads (or to only do it under absolute necessity)
I'm also curious to know if Waymo's are allowed to drop you slightly off from the address you specified -- in which case they can drop you at the next corner.
>I'm also curious to know if Waymo's are allowed to drop you slightly off from the address you specified -- in which case they can drop you at the next corner.
Yes, they do.
>Tell us where you want to go
>Choose your destination and we’ll select the safest spots to pick you up and drop you off.
Yes but thats a hack, the issue that it can’t handle those situations is still there and in my experience the car behind me was a waymo that kept trying to center but as soon as it saw cars coming it swerved. It kept going left and right. Unsafe in that situation and scared for people who were coming its way.
Why would they handle it? The autonomous car has all the time in the world. If it takes too long, the car can just hit an error condition, terminate the ride, or maybe tell passengers to walk to the end of the street to get into another car. Then it can just ignore its surroundings and wait things out.
These cars get away with being incapable of following police instructions, I don't see why they would need to care about other traffic users.
Around Portobello Road you often get some quite impressive jams like that with two busses and about ten cars. If it just stopped in the middle it wouldn't be popular with the permitting folk.
In Waymo's favour the humans often don't do that good a job.
I would love to see this navigating central London on a Saturday night. It is a mind draining nightmare.
Also, the small streets which are one car wide, and where one often needs to look far ahead to see whether there's a gap for you to sneak into whilst letting other cars by, will also be good to see handled.
They might so as they did elsewhere like choose few streets and saying they operate driverless in London just for show. I don't think one can take their word until it is done.
If they can operate in San Francisco, eg North Beach, they can operate in London. Likely they will just avoid narrow roads for a while. But I was very impressed in SF to see that Waymos are quite pushy to enter / cross traffic.
The only people who can afford daily taxi, twice a day, are already rich. Those are the people who can afford any private car parking price near the office, or most likely will have a spot reserved included in their high paying job.
I do agree that this fleet will slightly alleviate parking issue for people sporadically taking cars, not regularly. But at the cost of more congestion on the roads, especially in the already congested city centers, since that will be overwhelming majority of trip start or end points.
Quite the opposite IMHO. This helps reduce people who would hypothetically drunk drive on a Saturday evening, which in turn decreases the possibility that someone dies because of that (either the driver or a victim that was just passing / driving by).
Tbh, the sooner we remove the human from the equation, the better. It's scary to think that we allow so many careless people to drive vehicles that can kill people. I'm not talking just about drunk driving, but all the sort of distractions (smartphone, looking somewhere else, ...).
London specifically, AFAIK after midnight has no tube service. This means that Waymo (or whoever takes a similar initiative) actually helps towards creating a public transportation service that is cheaper and even safer than the current one. I'm personally all up for it - don't tax innovation!
> This helps reduce people who would hypothetically drunk drive on a Saturday evening
This was solved by taxis, and now uber, decades ago. If you're dumb enough to drive under influence in 2025 the cure isn't a driverless taxi it's 10 years in jail.
Sadly jail time doesn't often come to those who do it (at least, from experience / hearing stories of intoxicated drivers) and the consequences are paid by those who aren't intoxicated (e.g: getting killed by a drunk driver).
It's definitely not a cure, but removing the human factor (aside from the intoxication part) is anyways a very good goal IMHO.
Oh and btw, I've seen also taxi / uber drivers that were under the influence of alcohol / cocaine. Humans are the problem.
One argument would be that a driver in a cab will pay tax, a robot taxi will pay a lot less. That is quite a lot of money that is funneled to private companies instead of being used to improve our infrastructure.
Wow I thought this would take at least another decade, given how difficult driving in London is compared to American cities. I will still be really surprised if they can actually make this work
When I visited San Francisco recently the Waymos were really awesome and worked well, but also there's barely any traffic compared to London. The streets are all really wide and you can pretty much just pull over anywhere. Some even just stopped in the middle of the road and I was amazed to see people waiting patiently behind them! London is entirely different.
Still, props for trying. Will be very interesting to watch what happens!
Even in America they've just chosen the easier cities. I would love it if Waymo came to Boston but until it learns to proactively break traffic laws like all the uber drivers I'm not sure it's going to be very competitive :/.
I mean, some drivers do drive dangerously, but that's not what I'm talking about. With the narrow network of one way roads, going down an alley or going the wrong way down an empty one way road for 20 feet can sometimes make a 15 to 30 minute difference in trip time. And in traffic, it can be hard to get across some intersections unless you block the intersection. None of these is dangerous behaviors, per se, but they can be difference makers for riders.
Waymo is coming to DC and London. I assume they'll also be targeting centers of political power throughout the rest of the world. Starting on the West Coast made sense as a technical strategy, but the huge disadvantage is that nationa legislators and regulators don't go there enough. So lots of people can blithely assume that self-driving taxis are a decade away, or that Elon Musk is going to solve it. That's a dangerous business problem. Waymos don't seem possible til you've ridden a few of them, and then they're ordinary.
I’ve been to London a lot over the years and I’ve noticed black cabs have changed the last couple years. It used to always be a native Londoner who passed the knowledge and got access to driving black cab. But recently I’ve had some black cabs with immigrants - this would never be the case historically. So I think they all see the writing on the wall here.
The cab drivers are still true londoners who have passed the knowledge.
What has changed is how acceptable it is to be racist on hacker news and conflate native-londoner with whiteness. Or to conflate being being a true londoner with this xenophobic concept of nativeness.
So many of us londoners are non-white with London native parents who are non-white and in London that doesn't get questioned. You can't imagine how much it hurts to have outsiders come and try and tell us we are different and don't belong just because of our skin colour.
I think they’ve really let up on it recently. Have had a few rides where the driver was clueless and had to punch it into a GPS. Never happened to me previously.
> before it becomes stable and measurably successful; nobody dies
It’s already there for non-freeway driving. (Nobody dying is a perfectionist metric. It’s better than humans.)
What’s limiting it are capital costs. Once Waymo finds a non-Jaguar form factor it can mass manufacture, I imagine this would get rolled out rapidly. (To the extent Tesla has a shot, it’s in its mass manufacturing expertise.)
> Why is freeway driving more difficult (genuine question)?
“Stopping in lane becomes much more dangerous with the possibility of a rear-end collision at high speed. All stopping should be planned well in advance, ideally exiting at the next ramp, or at least driving to the closest shoulder with enough room to park.
This greatly increases the scope of edge cases that need to be handled autonomously and at freeway speeds.
…
The features that make freeways simpler — controlled access, no intersections, one-way traffic — also make ‘interesting’ events more rare. This is a double-edged sword. While the simpler environment reduces the number of software features to be developed, it also increases the iteration time and cost.
During development, ‘interesting’ events are needed to train data-hungry ML models. For validation, each new software version to be qualified for driverless operation needs to encounter a minimum number of ‘interesting’ events before comparisons to a human safety level can have statistical significance. Overall, iteration becomes more expensive when it takes more vehicle-hours to collect each event.”
> Would they delay their rollout over the capital costs of a few thousand $60,000 Jaguars?
Yes. Google, better than many others, gets scaling economics.
(And the cost is likely $100+ thousand per vehicle, after sensors and compute, with costs rising rapidly if you try and order too many too fast due to supply-chain bottlenecks nobody has bothered optimizing yet.)
Surely they need the same sensors and compute no matter what car they attach them to - they're either strapping $75k of sensors to a $60k luxury EV, or a $30k economy EV.
Unless they're planning 20mph golf carts to save on long-distance sensors, which as far as I know they aren't.
IMHO it's more likely they have capacity constraints because of all the different parts involved in a roll-out; you can't just double the capacity of a production facility without hiring lots of inexperienced new hires, meaning they can only build the things so fast. You can't roll out to a new city without places to park the cars, and places to charge them, and people to repair them, and places to store the spare parts, and trade routes to replenish the spare parts, and so on.
> Surely they need the same sensors and compute no matter what car they attach them to
Sure. But they may not have decided precisely what mix they need where, yet. More importantly, those suppliers may not have invested in factories that can produce millions of Waymos a year.
Ships, like planes, are fairly autonomous today and usually only have a very small crew. There are however multiple good reasons you don't want to have fully autonomous ships. They operate continuously in salt water for long duration away from ports, the maps they have are far from perfect and the sea bottom shift unpredictable, GPS are not always reliable, and broken down ship at sea is a massive safety and environmental risk.
It should not be understated how troublesome sea water is to complex machinery.
Because of the responsibility. In my country, if you kill someone on the road, you go to jail, and pay a huge fine. Which Google exec is going to jail if Waymo kill someone?
You prefer more dead people and more people in prison, rather than fewer of both?
Also, the same event (e.g. someone dying in a car crash) doesn't always have the same responsibilities behind it. If I kill someone by driving recklessly, I have more responsibility than if I kill someone when a bird crashes on my windshield. There are extreme cases where someone bears full responsibility, and extreme cases where an accident is just an accident and nobody is responsible. It may be that with self driving, a larger percentage of cases lean on the "true accident" side. (It's just an idea though, I agree there's an important question here that merits careful consideration.)
"Nobody dies" is the metric for as long as companies are not held responsible for their self-driving cars the same way people are. People are fallible, but for that reason they are also held responsible. If your company cannot be held responsible, it must not be fallible. And no, a court settlement is not being held responsible, that's just paying your way out of the justice system.
reply