Waymo

[mark95476]

Waymo has only been working on ONE thing and they've been at it for many years. They're bankrolled by Google so they effectively have unlimited money to throw at these problems. I would have thought they'd be demonstrating their superiority and reminding Elon/Tesla, and the rest of the industry, of it every single day. They're not. One of their top guys (got pardoned recently) took their technology to Uber, made a lot of money, but that effort wound up being sold off to Aurora.

I get the feeling that the prospects for Waymo actually making money is getting more improbable and maybe the board wants the bleeding to slow or stop.

Honestly, I think I would prefer if the co-CEO's stayed away from Twitter feuds and just focused on making Waymo's autonomous driving even better and expanding Waymo to more areas. Results speak louder than tweets.

 

[diplomat33]

Waymo has only been working on ONE thing and they've been at it for many years.

Yes, because contrary to what Elon claims, real driverless autonomous driving that is safe and reliable is very difficult and takes time.

I would have thought they'd be demonstrating their superiority and reminding Elon/Tesla, and the rest of the industry, of it every single day. They're not.

By several metrics, I think it could be argued that Waymo has already shown their FSD superiority over Tesla's FSD. They have the biggest actual driverless robotaxi ride-hailing service that the public can use. Tesla has no robotaxi ride-hailing fleet. They have the most autonomous miles. Waymo reported 20M autonomous miles, Tesla has only reported 12 autonomous miles. And Waymo has reported the best disengagement rate. Tesla won't even report disengagement data and admits they only have L2 which is not autonomous driving. Of course, Waymo is never going to convince the diehard Tesla fanboys that their FSD is better than Tesla's FSD. So I think spending energy trying to "rub it in" to Tesla fans is silly. Again, it makes more sense for Waymo to keep their eye on the ball, and focus on developing better FSD and expanding to more areas.

I get the feeling that the prospects for Waymo actually making money is getting more improbable and maybe the board wants the bleeding to slow or stop.

I think it is premature to assume that Waymo can't be profitable. There are companies who take years before they achieve profitability (Amazon, Tesla).

 

[diplomat33]

Waymo now let's you play any music from your phone through the car's speakers:

https://twitter.com/x/status/1379169894311268360

 

[DanCar]

... maybe the board wants the bleeding to slow or stop.

Yes, that has been true for quite some time. The bleeding has stopped. Waymo is now funding by selling parts of the company off to investors, in funding rounds.

 

[mark95476]

I think it was Waymo's former CEO that tried to convince Elon/Tesla. He's the very definition of a Tesla fanboy!

Of course, Waymo is never going to convince the diehard Tesla fanboys that their FSD is better than Tesla's FSD. So I think spending energy trying to "rub it in" to Tesla fans is silly. Again, it makes more sense for Waymo to keep their eye on the ball, and focus on developing better FSD and expanding to more areas.

 

[diplomat33]

I think it was Waymo's former CEO that tried to convince Elon/Tesla. He's the very definition of a Tesla fanboy!

What?! Elon started the feud with Waymo. Krafcik simply responded.

 

[Doggydogworld]

Krafcik was hired for his auto industry connections. In the long term he was to forge relationships with carmakers to embed the Waymo Driver into consumer cars, similar to how phones embed Android and browsers/web sites embed Google search. In the near term he was to start a Robotaxi business.

Two problems. First, they need an entrepreneur to get Robotaxi going and auto industry execs are famously not entrepreneurial. Second, the last thing carmakers want is to become "Android phone makers", so that was a tough sell from day one.

Krafcik couldn't even nail down the first "easy" phase -- convincing carmakers to build RoboPods. His failure to follow through on Phase 1A -- ordering 62k Pacificas and 20k i-Paces -- was IMHO the final straw. It shattered his credibility with OEMs, and since OEM relationships was all he really brought to the table he had to go.

Waymo must now decide whether to fully pursue Robotaxis or not. It's a huge gamble. If you don't serve an entire city you're useless. Consider @JJRicks who is possibly Waymo's top customer. He only really uses it to make videos because it doesn't go where he needs. Serving an entire city with reasonable wait times requires $3-5 billion in rolling stock plus another billion or so in ramp up losses. If you fail you're toast. If you succeed congratulations, you now own one city. Only 499 to go! Each with competitors frantically staking claims.

Trucking is a vastly better model. A single Robotruck is useful, a fleet of a hundred is a viable business. Sensors are a small fraction of Semi truck capex, and you dispense with expensive drivers who by law must shut down half the time. You can make money almost from day one, with a much smaller initial investment. Trucking is a regulatory game, though. Note that new co-CEO Tekedra Mawakana is a lawyer with an extensive regulatory and public policy background. She's the last person I'd hire to start to lead a startup, but the first I'd hire to enter a heavily regulated business.

It's easy to see this move as abandoning the Robotaxi business model in favor of trucking. I'm not quite ready to go that far yet, but the tea leaves keep lining up.

There are less risky ways to pursue Robotaxi, but Waymo has shown zero interest so far. Maybe that will change under Dol-Mawa. Or maybe they're getting ready to bet big after all. If they do I believe Magna will build the Pods. That's a much better strategic fit.

 

[diplomat33]

Krafcik was hired for his auto industry connections. In the long term he was to forge relationships with carmakers to embed the Waymo Driver into consumer cars, similar to how phones embed Android and browsers/web sites embed Google search. In the near term he was to start a Robotaxi business.

Two problems. First, they need an entrepreneur to get Robotaxi going and auto industry execs are famously not entrepreneurial. Second, the last thing carmakers want is to become "Android phone makers", so that was a tough sell from day one.

Krafcik couldn't even nail down the first "easy" phase -- convincing carmakers to build RoboPods. His failure to follow through on Phase 1A -- ordering 62k Pacificas and 20k i-Paces -- was IMHO the final straw. It shattered his credibility with OEMs, and since OEM relationships was all he really brought to the table he had to go.

Waymo must now decide whether to fully pursue Robotaxis or not. It's a huge gamble. If you don't serve an entire city you're useless. Consider @JJRicks who is possibly Waymo's top customer. He only really uses it to make videos because it doesn't go where he needs. Serving an entire city with reasonable wait times requires $3-5 billion in rolling stock plus another billion or so in ramp up losses. If you fail you're toast. If you succeed congratulations, you now own one city. Only 499 to go! Each with competitors frantically staking claims.

Trucking is a vastly better model. A single Robotruck is useful, a fleet of a hundred is a viable business. Sensors are a small fraction of Semi truck capex, and you dispense with expensive drivers who by law must shut down half the time. You can make money almost from day one, with a much smaller initial investment. Trucking is a regulatory game, though. Note that new co-CEO Tekedra Mawakana is a lawyer with an extensive regulatory and public policy background. She's the last person I'd hire to start to lead a startup, but the first I'd hire to enter a heavily regulated business.

It's easy to see this move as abandoning the Robotaxi business model in favor of trucking. I'm not quite ready to go that far yet, but the tea leaves keep lining up.

There are less risky ways to pursue Robotaxi, but Waymo has shown zero interest so far. Maybe that will change under Dol-Mawa. Or maybe they're getting ready to bet big after all. If they do I believe Magna will build the Pods. That's a much better strategic fit.

I don't agree that if you don't serve an entire city, you are useless. It's not like JJ Ricks is Waymo's only customer or that all of Waymo's ride-hailing customers just use the service to make videos. There are people who use the service every month to actually get to places they need. So it is not completely useless, just less useful. But the smaller the service area, the less useful it is, because it will serve fewer people. So yes, you want the biggest service area possible, in order to serve as many people as possible. So Waymo certainly wants to service the biggest area possible in order to reach as many customers as possible.

I agree that the self-driving truck business is potentially lucrative. So far though, Waymo has not said anything about abandoning Waymo One. Rather, Waymo has indicated that they want to do both Waymo One and Waymo Via. We will see if that changes.

As I see it, Waymo has a few different paths:
1) Expand Waymo One to bigger areas & grow Waymo Via.
2) Launch personal car ownership where people can lease a robotaxi.
3) Abandon Waymo One and focus on Waymo Via.

It will be interesting to see what direction Dolgov and Mawakana choose.

I do think Waymo has basically 2 challenges:
1) Achieve a profitable business model.
Waymo claims that they have a path to profitability with the 5th Generation hardware that only costs $0.30/mile. But so far, I don't think Waymo has actually shown profitability. So that is a challenge.
2) Prove their FSD is safe in a large area.
Waymo has shown some encouraging data that their autonomous driving is safe in the Phoenix area. But proving safety on a larger scale is harder.

If Waymo can do both these things, they have a clear path forward.

 

[diplomat33]

Waymo is taking questions on Twitter:

https://twitter.com/x/status/1380241538198933512

 

[diplomat33]

Waymo answered one of my questions!

https://twitter.com/x/status/1380578382790938637

Interesting that they added more cameras to provide more overlap and more angles.

 

[powertoold]

Less techy guy reviews two Waymo rides, very positive

 

[diplomat33]

Less techy guy reviews two Waymo rides, very positive

I get that he is not a tech guy but his description of the Waymo sensors was very lacking. He completely forgets to mention that the Waymo uses cameras. He only mentions radar and lidar. And the graphic of the sensors was incomplete.

 

[diplomat33]

00:00 Car arriving and R.I.P. balloon
00:20 Boarding
00:38 Ride start and rider support call
01:00 Parking lot/pedestrians
01:55 Unprotected right
02:15 Cyclist
02:48 Unprotected left
04:25 Unprotected right
07:26 Recentering on a road with no lane lines
07:38 Technically an unprotected left but not really exciting
08:20 Edging forward on unprotected right until surroundings are fully visible (watch the planner route)
08:39 Technically an unprotected left but not really exciting
09:40 Unprotected right and Spotify issues
10:25 Person honks at me
11:25 Major Spotify issues
12:10 Music doesn't work on Spotify, but PODCASTS do????
12:35 Protected right
12:50 Goes around stopped bus, close but don't touch

Overall, the Waymo handles the drive really well IMO.

At 10:25 the Waymo does get honked at for not making a right turn quick enough. Kinda funny to honk at a driverless car. This is probably a good example of the type of cases where Waymo can be overly cautious to the point of annoying other drivers. In fact, I might compare Waymo to a new driver that is competent but also overly cautious at times. IMO, this is something that Waymo can work on, to make the driving policy more confident and more assertive, while still being safe.

 

[diplomat33]

"Waymo’s new co-CEOs Tekedra Mawakana and Dmitri Dolgov say Alphabet’s self-driving powerhouse is sticking to a safety-conscious roadmap to robotic ride-hailing, trucks and personal vehicles, even if that means a longer path to commercialization. In their first joint interview, they also share concerns that scrutiny of fatal Tesla accidents, related to its partially automated system, could chip away at consumer confidence in autonomous technology. "

Read more here:

Amid Tesla Crash Concerns Waymo’s New Co-CEOs Say No Shortcuts To Safe Autonomy

Alphabet’s self-driving powerhouse is sticking with a safety-conscious roadmap to robotic ride-hailing, trucks and personal vehicles, even if that means a longer path to commercialization.

www.forbes.com

 

[diplomat33]

Waymo will be doing a reddit AMA on Friday from 1:30-2:30pm PT:

https://www.reddit.com/r/SelfDrivingCars/comments/n031vq

 

[Doggydogworld]

They should rename these to Ask Me Anything That Doesn't Matter They won't answer questions people care about, like timeline for expanding in Phoenix and/or other cities. They won't tell us anything about costs or operating metrics, e.g. monitor per vehicle ratio, deadhead ratio, etc., or improvements needed before expansion. They won't talk about how they see the market evolving, e.g. Uber replacement, high value route (e.g. airport) takeover, second car replacement, etc.

This is all stuff entrepreneurs regularly discuss. They may withhold certain details, but almost all will give you their view of the market and their plan and timeline. They do this in part to get customers and potential partners excited, but also to get valuable feedback. Waymo is like a sealed-off room, though, with little air exchange. That made sense when they were an R&D project, but they spun out and became a "business" over 5 years ago. When will they start acting like one?

Don't look to the co-CEOs to change anything. Neither is the least bit entrepreneurial.

 

[diplomat33]

They should rename these to Ask Me Anything That Doesn't Matter They won't answer questions people care about, like timeline for expanding in Phoenix and/or other cities. They won't tell us anything about costs or operating metrics, e.g. monitor per vehicle ratio, deadhead ratio, etc., or improvements needed before expansion. They won't talk about how they see the market evolving, e.g. Uber replacement, high value route (e.g. airport) takeover, second car replacement, etc.

This is all stuff entrepreneurs regularly discuss. They may withhold certain details, but almost all will give you their view of the market and their plan and timeline. They do this in part to get customers and potential partners excited, but also to get valuable feedback. Waymo is like a sealed-off room, though, with little air exchange. That made sense when they were an R&D project, but they spun out and became a "business" over 5 years ago. When will they start acting like one?

Don't look to the co-CEOs to change anything. Neither is the least bit entrepreneurial.

Uh?! Several people asked about timelines for expansion and scaling up to other cities. And the questions got a lot of upvotes. I think they will answer them. Let's see what they say before judging and dismissing it.

 

[Doggydogworld]

Uh?! Several people asked about timelines for expansion and scaling up to other cities. And the questions got a lot of upvotes. I think they will answer them. Let's see what they say before judging and dismissing it.

They've done AMAs (or as I call them AMATDMs) in the past and evaded these same questions. Heck, they've had these questions on their FAQ page for 3+ years. The non-answers never change.

Entrepreneurs are reckless. Sometimes criminally so, unless they listen to sane voices around them. Levandowski was probably their only entrepreneur. Not that I'd endorse him for CEO, but I don't see how they'll shift out of PARC without someone of his ilk.

 

[diplomat33]

They've done AMAs (or as I call them AMATDMs) in the past and evaded these same questions. Heck, they've had these questions on their FAQ page for 3+ years. The non-answers never change.

Entrepreneurs are reckless. Sometimes criminally so, unless they listen to sane voices around them. Levandowski was probably their only entrepreneur. Not that I'd endorse him for CEO, but I don't see how they'll shift out of PARC without someone of his ilk.

If Levandowski is your idea of the kind of leadership Waymo should have, then count me out. Levandowski got 18 months in prison for stealing secrets. That's not exactly what you want in your CEO. I'm glad Waymo did not go with him or someone like him to lead the company.

I get that you think Waymo needs more of an entrepreneurial spirit in order to expand faster. And I agree that you don't be so timid that you never expand your business. But you also don't want to be reckless where your deploy 1M robotaxis before they are ready and your robotaxis get into a bunch of accidents and your company gets sued for billions or worse your CEO goes to prison.

 

[EVNow]

worse your CEO goes to prison.

Thats all they care about - if so - they are not suitable to be CEOs.

Its been said CEOs are borderline psychopaths - I'd not endorse such people. But people who are too careful won't do well in totally new spaces like EVs or rockets or robotaxis. Ironically silicon valley is full of brave CEOs ... just that you can't find them at Waymo.

Why (Some) Psychopaths Make Great CEOs

British journalist Jon Ronson immersed himself in the world of mental health diagnosis and criminal profiling to understand what makes some people psychopaths -- dangerous predators who lack the behavioral controls and tender feelings the rest of us take for granted. Among the things he learned...

www.forbes.com

ps : Basically you need people who push boundaries - take considerable risks. Unfortunately a lot of them push boundaries in all spheres - social, ethical - not just technical.