I know these things aren't perfect, but self-driving cars is one of the few upcoming technologies I'm actually been looking forward to for a while. I hate actually driving, but I like everything...
I know these things aren't perfect, but self-driving cars is one of the few upcoming technologies I'm actually been looking forward to for a while. I hate actually driving, but I like everything else about personal vehicle travel.
I used to be a lot more excited about them. It's become clear to me that EVs and self driving cars have become a useful tool for people who have always and will always oppose public transit, which...
I used to be a lot more excited about them. It's become clear to me that EVs and self driving cars have become a useful tool for people who have always and will always oppose public transit, which has always been our best tool for combating climate change in the transportation sector, and our best bet in combatting the urban development antipatterns that have so severely harmed north American society in general over the past 80 years.
I'm not convinced that either technology will significantly address either of those issues, and have the potential to encourage private vehicle ownership even more than it is now, just exacerbating those problems in the long term.
well, they can, but won't. For very rough napkin math: LA has about 2500 buses. Imagine a magnitude more of those buses (maybe, 20k buses with 50 seats each) that run 24/7 with very low wait times...
well, they can, but won't. For very rough napkin math:
LA has about 2500 buses. Imagine a magnitude more of those buses (maybe, 20k buses with 50 seats each) that run 24/7 with very low wait times across 90%+ of the city. That'd be up to 100k more seats to deploy, cutting a good deal into the 300k cars on the freeway on any given day. without changing the structure of the city, you can potentially reduce traffic by 25% and be more efficient with bus routes than ever. All that is even possible today, but the allure of not having to pay 15k more bus drivers could push that needle.
But that likely won't happen for more political reasons. Car industry doesn't want to take 100k more cars off the road, the US public has a lot of desire for private transportation due to convinience, and there's still a lot of underlying mental health and homeless issues that makes public transportation less appealing or even less safe. These aren't issues solved by technology.
It doesn’t have to either or, at least with self driving. In the US, often you must use Uber to bridge the gap between public transit gaps. Uber existing makes public transit only much more viable.
It doesn’t have to either or, at least with self driving. In the US, often you must use Uber to bridge the gap between public transit gaps. Uber existing makes public transit only much more viable.
You're probably right. I have no desire to live in an urban setting where public transportation can thrive. I'm not against the ideas or anything, it's just not for me. So I'm excited about the...
You're probably right. I have no desire to live in an urban setting where public transportation can thrive. I'm not against the ideas or anything, it's just not for me. So I'm excited about the continued benefits of personal over public transportation combined with claiming some of that time back to do something else while the vehicle is moving.
Urban settings aren't the only place where public transit can thrive. There are a lot of small towns, mostly in Europe, that have robust public transit and where cars aren't used for most trips....
Urban settings aren't the only place where public transit can thrive. There are a lot of small towns, mostly in Europe, that have robust public transit and where cars aren't used for most trips. The problem is that with north American development patterns, you're right, public transit isn't workable. Even the smallest towns are designed primarily for people to get around with cars. It doesn't necessarily have to be that way though, and in fact it can't continue to be that way, if we want to avoid the worst effects of climate change. Even EVs use far too much energy and materials to produce and operate.
The paradigm of every human adult owning a 4000 pound chunk of steel and plastic that they throw away and replace every five years just isn't sustainable no matter what those chunks are powered by. It's the primary reason why humanity has been able to do such an ungodly amount of damage to the planet in 80 years despite living in relative harmony with the global ecosystem for the previpus 300,000.
Sure, but I purposely live outside of urban settings because I want a SFH on a decent chunk of land accompanied by quiet and privacy. I'm probably in the minority on this website (but perhaps not...
Sure, but I purposely live outside of urban settings because I want a SFH on a decent chunk of land accompanied by quiet and privacy. I'm probably in the minority on this website (but perhaps not so much across the general public) that actually likes a suburban setting. I'll reduce my carbon footprint elsewhere and keep my preferred living style, which requires a personal vehicle. I'd be happy for public transportation to pick me up somewhere within walking distance of my home but that's certainly not sustainable at the density (or lack thereof) of my preferred living style.
The average age of a passenger car in the US is 13 years, and rising. (People don't usually own them that long, but they sell them to someone else.) I couldn't find good data for how long cars...
(People don't usually own them that long, but they sell them to someone else.)
I couldn't find good data for how long cars last before they're no longer used, but it will be more than that.
I expect that one reason that people hang onto cars longer is that both new and used cars are more expensive than they used to be. But maybe it works the other way, too? The longer they're expected to last, the more they're worth.
This is surprising since it has a really bad reputation in SF. This article gives some good examples of the dangerous things they still do. It has driven people to mess with and even destroy them.
This is surprising since it has a really bad reputation in SF. This article gives some good examples of the dangerous things they still do. It has driven people to mess with and even destroy them.
I think they shouldn't have started in SF. I understand why: that's where a lot of bleeding edge tech companies are. However, from what I understand, the bay area has reasonable public...
I think they shouldn't have started in SF. I understand why: that's where a lot of bleeding edge tech companies are. However, from what I understand, the bay area has reasonable public transportation and walkable areas, and it's dense and congested.
You know where none of that exists? The hellish car-centric landscape of Midwest suburbs. I live in the densest metro area in my state, and there is basically no practical public transportation (only buses, which take hours to go anywhere). There are few pedestrians because the area is not walkable (downright scary at times, many people no longer yield to pedestrians at crossing). Evil Knievel would not dare to ride a bicycle on the main streets here.
Unless we raze the suburbs and force millions to move into scarce and unaffordable city housing, we will need cars for the foreseeable future out here. This car free city thing just won't work in the Midwest suburb. SF might not want AVs, but we need something out here. Uber was a game changer for the suburbs. AVs could be a realistic alternative to car ownership in this car nightmare.
Honestly, it's not. SF is the most dense, but it's not all that dense. The vast majority of the city is zoned for single home families. I remember when I use to go to Berkeley and visit SF,...
Honestly, it's not. SF is the most dense, but it's not all that dense. The vast majority of the city is zoned for single home families. I remember when I use to go to Berkeley and visit SF, sometimes I wouldn't be able to distinguish between the suburbs (note, like 70% of the city) of SF and Berkeley. There's only a handful of areas that resemble anything like density. Public transit is very hit and miss. If you happen to be near a MUNI line, it's not horrible, although the way the lines are designed are really stupid. Or I guess if you live near BART but that's a whole other can of worms. If you don't, it's pretty garbage. It's one of the cities where if you swap between "car" and "public transit" on google maps, the trip duration goes up by an hour.
I know a lot of people without cars in SF, but that's more from practical reasons (getting a parking spot with your apartment can be +$600/month!), and they just Uber a lot. So in that respect, if they can replace all the Ubers with self driving Uber-equivalents, that's something.
And the moment you get out of SF, forget about anything except single story SFHs. South bay is absolutely a "car centric hellscape", if you want to describe it like that. Buses are far and few between, suck, and the only rail is VTA, which also sucks and only has one line in San Jose. If you don't have a car, you're just screwed. Or just Uber everywhere.
I agree with @scottc though. Perhaps they should have chosen San Jose over San Francisco rather than the midwest with its ice and snow. However San Francisco has frequent street fairs and...
I agree with @scottc though. Perhaps they should have chosen San Jose over San Francisco rather than the midwest with its ice and snow. However San Francisco has frequent street fairs and festivals and public holidays where people take over the streets. The autonomous car that got set on fire was trying to drive through crowds in Chinatown on chinese new year. San Francisco streets are narrow enough and population is dense enough that a confused self driving vehicle that stops while it figures out what to do is a serious traffic impediment and frustration to many people. Likewise driveways are so close together that a stopped self driving vehicle that pulls over is likely to block one and there have been complaints that they block driveways for long periods of time.
San Jose or Fremont or Concord would have been a much better choice and the cars would have irritated fewer people and been a lot less noticeable.
I wouldn't suggest self driving vehicles in San Francisco and I wouldn't suggest them in New Orleans, for similar reasons
From the article: Supposedly the coverage maps are here, but I don't know where they got the maps from.
From the article:
Waymo received approval Friday afternoon from the California Public Utilities Commission to operate a commercial robotaxi service in Los Angeles, the San Francisco Peninsula and on San Francisco freeways.
The approval removes the last barrier for the Alphabet company to charge for rides in these expanded areas. Importantly, it opens up new territory for Waymo in one of the country’s largest cities and unlocks a route to San Francisco International Airport, which is located south of the city.
Supposedly the coverage maps are here, but I don't know where they got the maps from.
I’d like to be wrong but I see almost all of this like Uber all over again. People bending over backwards to ignore companies that skip regulations or aren’t under enough scrutiny because it’s...
I’d like to be wrong but I see almost all of this like Uber all over again. People bending over backwards to ignore companies that skip regulations or aren’t under enough scrutiny because it’s convenient now.
I really don’t look forward to more adoption of this stuff with current behaviors and especially once they capture enough of the market to cut corners on quality
It wasn’t because of one bad accident. It was because an at first unavoidable accident occurred, then it continued driving with someone under it, and then they tried to lie about it. It bothers me...
It wasn’t because of one bad accident.
It was because an at first unavoidable accident occurred, then it continued driving with someone under it, and then they tried to lie about it.
It bothers me that such a product, by such a company, was on the road at all.
And again, just because they’re good now when all eyes are on them doesn’t mean they will be the moment they have captured the market. People were screaming about how great Uber was instead of taxis and now it’s got much the same issues.
Worth noting- this incident was under Cruise (GM), not Waymo (Google). https://arstechnica.com/tech-policy/2023/11/after-robotaxi-dragged-pedestrian-20-feet-cruise-founder-and-ceo-resigns/
Worth noting- this incident was under Cruise (GM), not Waymo (Google).
It does seem that it's not just the accident itself, but how they handled it. There was bad decision-making about how to cooperate with regulators that got Cruise in the end. They half-assed it...
It does seem that it's not just the accident itself, but how they handled it. There was bad decision-making about how to cooperate with regulators that got Cruise in the end. They half-assed it with a passive "let the video speak for itself" approach and showed they weren't serious.
If you want a detailed account of how it went down, this blog post is pretty good:
On reading the entire report, I'd say that Cruise both looks better and worse than in the articles I saw, which is the same pattern we saw when we looked at the actual source for Exhibits H and J from Twitter v. Musk, the United States v. Microsoft Corp. docs, etc.; just as some journalists seem to be pro/anti-Elon Musk and pro/anti-Microsoft, willing to push an inaccurate narrative to dunk on them to the maximum extent possible or exonerate them to the maximum extent possible, we see the same thing here with Cruise. And as we saw in those cases, despite some articles seemingly trying to paint Cruise in the best or worst light possible, the report itself has material that is more positive and more negative than we see in the most positive or negative stories.
Aside from correcting misleading opinions on the report, I find the report interesting because it's rare to see any kind of investigation over what went wrong in tech in this level of detail, let alone a public one. We often see this kind of investigation in safety critical systems and sometimes see in sports as well as for historical events, but tech events are usually not covered like this. Of course companies do post-mortems of incidents, but you generally won't see a 200-page report on a single incident, nor will the focus of post-mortems be what the focus was here. In the past, we've noted that a lot can be learned by looking at the literature and incident reports on safety-critical systems, so of course this is true here as well, where we see a safety-critical system that's more tech adjacent than the ones we've looked at previously.
I've only skimmed, but noticed this bit in particular from the conclusions:
By the time Cruise employees from legal, government affairs, operations, and systems integrity met with regulators and other government officials on October 3, they knew or should have known that the Cruise AV had engaged in a pullover maneuver and dragged the pedestrian underneath the vehicle for approximately 20 feet"
"Cruise’s passive, non-transparent approach to its disclosure obligations to its regulators reflects a basic misunderstanding of what regulatory authorities need to know and when they need to know it"
"Although neither Cruise nor Quinn Emanuel can definitively establish that NHTSA or DMV were shown the entirety of the Full Video, including the pullover maneuver and dragging, the weight of the evidence indicates that Cruise attempted to play the Full Video in these meetings; however, internet connectivity issues impeded or prevented these regulators from seeing the video clearly or fully."
"in the face of these internet connectivity issues that caused the video to freeze or black- or white-out, Cruise employees remained silent, failing to ensure that the regulators understood what they likely could not see – that the Cruise AV had moved forward again after the initial impact, dragging the pedestrian underneath the vehicle"
"Even if, as some Cruise employees stated, they were unaware of the pullover maneuver and pedestrian dragging at the time of certain regulatory briefings (which itself raises other concerns), Cruise leadership and other personnel were informed about the full details of the October 2 Accident during the day on October 3 and should have taken corrective action."
In addition to what you said, Cruise was already getting a lot of negative headlines before this incident, though none quite as severe. So in addition to their mishandling post pedestrian dragging...
In addition to what you said, Cruise was already getting a lot of negative headlines before this incident, though none quite as severe. So in addition to their mishandling post pedestrian dragging incident, I think there was also a build-up of sorts that amplified the public scrutiny of the pedestrian getting dragged as well as Cruise mishandling the reporting and sharing of information afterwards. If it had simply been a one-off thing, surely it still would have been a big deal, but it would potentially give them a little more grace without that build up of pressure in the lesser incidents beforehand.
I simply cannot buy their explanation of "internet connectivity issues" that happened to occur in such a way that most favored Cruise in terms of not revealing that their hardware/software resulted in a pedestrian that was already hit by their car then getting dragged by the car afterwards. The amount of public pressure and scrutiny Cruise was already receiving was quite high and makes it more reasonable to believe that Cruise would have incentive to try to downplay an even more egregious mistake. It's too much of a coincidence given the history. They also initially released statements that omitted this aspect of their role in the pedestrian's injuries, so the fact that the "internet connectivity issues" happened to result in omissions of elements of the video that aligned with their omissions in initial public statements further makes the coincidence hard to believe, unless they want to argue their keyboard suffered technical issues while they were typing up their initial statements and only on the parts where describing the pullover and pedestrian dragging. At least that way they'd be consistent in arguing that technical issues resulted in omissions of the truth.
There's also a potential bit of irony to it because part of the scrutiny of these autonomous vehicles is that they simply just stop where they are and block traffic when they run into unexpected situations with no clear defined answer. The pullover that this Cruise vehicle used which resulted in the pedestrian getting dragged could likely well have been in part a result of public pressure over how these vehicles react to unexpected situations. Absent public pressure, there's really almost no incentive at the current scale they operate at for autonomous vehicles to pull over to not block traffic. If all vehicles on the road were autonomous, then yeah maybe they would have other incentives because the vehicles stuck behind them would be their own that are losing time, and time is money. Yet as it is now, they're far more incentivized to just stop there and let everyone else deal with the fallout of a car stopped in the middle of the road and the public pressure of that is what sort of gives them incentive to try to balance public scrutiny and their own liabilities.
Although it does matter, I think you might be over-estimating the role of public pressure. The company and the regulators interact directly, and regulators make their own judgements based on that....
Although it does matter, I think you might be over-estimating the role of public pressure. The company and the regulators interact directly, and regulators make their own judgements based on that.
We can see it in this case where regulators approved Waymo’s expansion despite a lot of public sentiment against driverless cars by a public that often doesn’t distinguish between companies.
It would be interesting to know more about how they handle pull overs. Did Cruise change their algorithm? Did Cruise do it differently than Waymo? I don’t think reasoning from incentives is enough to answer these questions.
It's possible I'm overestimating it, but I think the issue in this case is that the regulators approving it are still representing the public, it's just not the local population in SF. So you...
It's possible I'm overestimating it, but I think the issue in this case is that the regulators approving it are still representing the public, it's just not the local population in SF. So you can't equate regulatory action with public sentiment unless you factor in who is being represented.
The problem with knowing how they handle things is that they control all the information about what incidents occur or nearly occur with these vehicles, and only report what they want to report or are required to report, but don't otherwise engage in much transparency to allow the public to understand the development of autonomous vehicles. Reasoning from incentive is simply what is left when there's a lack of transparency and inability to access detailed information about how the vehicles otherwise operate.
The representation is indirect and rather fuzzy. I assume regulators consider it their mission to protect the public, but that doesn’t mean they agree with any particular proposal about how to do...
The representation is indirect and rather fuzzy. I assume regulators consider it their mission to protect the public, but that doesn’t mean they agree with any particular proposal about how to do it, even if it’s popular. And yes, it’s a state agency, not a local one.
Sometimes we do get pretty detailed reports about what’s going on. Also, these services operate in public and other people report on what they do. Every accident makes the news. That’s one way regulatory agencies would find out if the company isn’t reporting things accurately to them (as happened with Cruise).
Yeah, I think that's the real kicker here. Accidents will still happen because no amount of sensors can defy physics, and early adopter technology will inevitably act in ways humans won't as they...
and then they tried to lie about it.
Yeah, I think that's the real kicker here. Accidents will still happen because no amount of sensors can defy physics, and early adopter technology will inevitably act in ways humans won't as they adapt to what humans expect.
But trying to more or less hide a life threatening event under PR because shareholders or whatever... it really shows where their priorities lie and why we need to regulate/sanction this behavior now rather than later.
Dumb probably was a poor choice of words, since I didn't mean it would be easy but just that it's surprising if no one with the expertise hasn't tried it. Obviously it would need initial seed...
Dumb probably was a poor choice of words, since I didn't mean it would be easy but just that it's surprising if no one with the expertise hasn't tried it.
Obviously it would need initial seed money and a lot of legwork since you wouldn't be marketing it directly to consumers. And it's not like the traditional taxi industry is doing that well, plus Silicon Valley probably favors the Uber model, as you can directly exploit the workers, when here there would be a person in the middle.
Still surprising that someone with the means hasn't done it (well).
I know these things aren't perfect, but self-driving cars is one of the few upcoming technologies I'm actually been looking forward to for a while. I hate actually driving, but I like everything else about personal vehicle travel.
I used to be a lot more excited about them. It's become clear to me that EVs and self driving cars have become a useful tool for people who have always and will always oppose public transit, which has always been our best tool for combating climate change in the transportation sector, and our best bet in combatting the urban development antipatterns that have so severely harmed north American society in general over the past 80 years.
I'm not convinced that either technology will significantly address either of those issues, and have the potential to encourage private vehicle ownership even more than it is now, just exacerbating those problems in the long term.
well, they can, but won't. For very rough napkin math:
LA has about 2500 buses. Imagine a magnitude more of those buses (maybe, 20k buses with 50 seats each) that run 24/7 with very low wait times across 90%+ of the city. That'd be up to 100k more seats to deploy, cutting a good deal into the 300k cars on the freeway on any given day. without changing the structure of the city, you can potentially reduce traffic by 25% and be more efficient with bus routes than ever. All that is even possible today, but the allure of not having to pay 15k more bus drivers could push that needle.
But that likely won't happen for more political reasons. Car industry doesn't want to take 100k more cars off the road, the US public has a lot of desire for private transportation due to convinience, and there's still a lot of underlying mental health and homeless issues that makes public transportation less appealing or even less safe. These aren't issues solved by technology.
It doesn’t have to either or, at least with self driving. In the US, often you must use Uber to bridge the gap between public transit gaps. Uber existing makes public transit only much more viable.
You're probably right. I have no desire to live in an urban setting where public transportation can thrive. I'm not against the ideas or anything, it's just not for me. So I'm excited about the continued benefits of personal over public transportation combined with claiming some of that time back to do something else while the vehicle is moving.
Urban settings aren't the only place where public transit can thrive. There are a lot of small towns, mostly in Europe, that have robust public transit and where cars aren't used for most trips. The problem is that with north American development patterns, you're right, public transit isn't workable. Even the smallest towns are designed primarily for people to get around with cars. It doesn't necessarily have to be that way though, and in fact it can't continue to be that way, if we want to avoid the worst effects of climate change. Even EVs use far too much energy and materials to produce and operate.
The paradigm of every human adult owning a 4000 pound chunk of steel and plastic that they throw away and replace every five years just isn't sustainable no matter what those chunks are powered by. It's the primary reason why humanity has been able to do such an ungodly amount of damage to the planet in 80 years despite living in relative harmony with the global ecosystem for the previpus 300,000.
Sure, but I purposely live outside of urban settings because I want a SFH on a decent chunk of land accompanied by quiet and privacy. I'm probably in the minority on this website (but perhaps not so much across the general public) that actually likes a suburban setting. I'll reduce my carbon footprint elsewhere and keep my preferred living style, which requires a personal vehicle. I'd be happy for public transportation to pick me up somewhere within walking distance of my home but that's certainly not sustainable at the density (or lack thereof) of my preferred living style.
The average age of a passenger car in the US is 13 years, and rising.
(People don't usually own them that long, but they sell them to someone else.)
I couldn't find good data for how long cars last before they're no longer used, but it will be more than that.
I expect that one reason that people hang onto cars longer is that both new and used cars are more expensive than they used to be. But maybe it works the other way, too? The longer they're expected to last, the more they're worth.
I have some pretty severe driving anxiety after some bad accidents earlier in my life. I'd really really love a self driving car.
This is surprising since it has a really bad reputation in SF. This article gives some good examples of the dangerous things they still do. It has driven people to mess with and even destroy them.
I think they shouldn't have started in SF. I understand why: that's where a lot of bleeding edge tech companies are. However, from what I understand, the bay area has reasonable public transportation and walkable areas, and it's dense and congested.
You know where none of that exists? The hellish car-centric landscape of Midwest suburbs. I live in the densest metro area in my state, and there is basically no practical public transportation (only buses, which take hours to go anywhere). There are few pedestrians because the area is not walkable (downright scary at times, many people no longer yield to pedestrians at crossing). Evil Knievel would not dare to ride a bicycle on the main streets here.
Unless we raze the suburbs and force millions to move into scarce and unaffordable city housing, we will need cars for the foreseeable future out here. This car free city thing just won't work in the Midwest suburb. SF might not want AVs, but we need something out here. Uber was a game changer for the suburbs. AVs could be a realistic alternative to car ownership in this car nightmare.
Honestly, it's not. SF is the most dense, but it's not all that dense. The vast majority of the city is zoned for single home families. I remember when I use to go to Berkeley and visit SF, sometimes I wouldn't be able to distinguish between the suburbs (note, like 70% of the city) of SF and Berkeley. There's only a handful of areas that resemble anything like density. Public transit is very hit and miss. If you happen to be near a MUNI line, it's not horrible, although the way the lines are designed are really stupid. Or I guess if you live near BART but that's a whole other can of worms. If you don't, it's pretty garbage. It's one of the cities where if you swap between "car" and "public transit" on google maps, the trip duration goes up by an hour.
I know a lot of people without cars in SF, but that's more from practical reasons (getting a parking spot with your apartment can be +$600/month!), and they just Uber a lot. So in that respect, if they can replace all the Ubers with self driving Uber-equivalents, that's something.
And the moment you get out of SF, forget about anything except single story SFHs. South bay is absolutely a "car centric hellscape", if you want to describe it like that. Buses are far and few between, suck, and the only rail is VTA, which also sucks and only has one line in San Jose. If you don't have a car, you're just screwed. Or just Uber everywhere.
I agree with @scottc though. Perhaps they should have chosen San Jose over San Francisco rather than the midwest with its ice and snow. However San Francisco has frequent street fairs and festivals and public holidays where people take over the streets. The autonomous car that got set on fire was trying to drive through crowds in Chinatown on chinese new year. San Francisco streets are narrow enough and population is dense enough that a confused self driving vehicle that stops while it figures out what to do is a serious traffic impediment and frustration to many people. Likewise driveways are so close together that a stopped self driving vehicle that pulls over is likely to block one and there have been complaints that they block driveways for long periods of time.
San Jose or Fremont or Concord would have been a much better choice and the cars would have irritated fewer people and been a lot less noticeable.
I wouldn't suggest self driving vehicles in San Francisco and I wouldn't suggest them in New Orleans, for similar reasons
From the article:
Supposedly the coverage maps are here, but I don't know where they got the maps from.
I’d like to be wrong but I see almost all of this like Uber all over again. People bending over backwards to ignore companies that skip regulations or aren’t under enough scrutiny because it’s convenient now.
I really don’t look forward to more adoption of this stuff with current behaviors and especially once they capture enough of the market to cut corners on quality
So far, they get cancelled after one bad accident. Waymo is still standing because they haven't had a bad enough accident yet.
It wasn’t because of one bad accident.
It was because an at first unavoidable accident occurred, then it continued driving with someone under it, and then they tried to lie about it.
It bothers me that such a product, by such a company, was on the road at all.
And again, just because they’re good now when all eyes are on them doesn’t mean they will be the moment they have captured the market. People were screaming about how great Uber was instead of taxis and now it’s got much the same issues.
Worth noting- this incident was under Cruise (GM), not Waymo (Google).
https://arstechnica.com/tech-policy/2023/11/after-robotaxi-dragged-pedestrian-20-feet-cruise-founder-and-ceo-resigns/
It does seem that it's not just the accident itself, but how they handled it. There was bad decision-making about how to cooperate with regulators that got Cruise in the end. They half-assed it with a passive "let the video speak for itself" approach and showed they weren't serious.
If you want a detailed account of how it went down, this blog post is pretty good:
He takes notes from this document: REPORT TO THE BOARDS OF DIRECTORS OF CRUISE LLC, GM CRUISE HOLDINGS LLC, AND GENERAL MOTORS HOLDINGS LLC REGARDING THE OCTOBER 2, 2023 ACCIDENT IN SAN FRANCISCO
I've only skimmed, but noticed this bit in particular from the conclusions:
In addition to what you said, Cruise was already getting a lot of negative headlines before this incident, though none quite as severe. So in addition to their mishandling post pedestrian dragging incident, I think there was also a build-up of sorts that amplified the public scrutiny of the pedestrian getting dragged as well as Cruise mishandling the reporting and sharing of information afterwards. If it had simply been a one-off thing, surely it still would have been a big deal, but it would potentially give them a little more grace without that build up of pressure in the lesser incidents beforehand.
I simply cannot buy their explanation of "internet connectivity issues" that happened to occur in such a way that most favored Cruise in terms of not revealing that their hardware/software resulted in a pedestrian that was already hit by their car then getting dragged by the car afterwards. The amount of public pressure and scrutiny Cruise was already receiving was quite high and makes it more reasonable to believe that Cruise would have incentive to try to downplay an even more egregious mistake. It's too much of a coincidence given the history. They also initially released statements that omitted this aspect of their role in the pedestrian's injuries, so the fact that the "internet connectivity issues" happened to result in omissions of elements of the video that aligned with their omissions in initial public statements further makes the coincidence hard to believe, unless they want to argue their keyboard suffered technical issues while they were typing up their initial statements and only on the parts where describing the pullover and pedestrian dragging. At least that way they'd be consistent in arguing that technical issues resulted in omissions of the truth.
There's also a potential bit of irony to it because part of the scrutiny of these autonomous vehicles is that they simply just stop where they are and block traffic when they run into unexpected situations with no clear defined answer. The pullover that this Cruise vehicle used which resulted in the pedestrian getting dragged could likely well have been in part a result of public pressure over how these vehicles react to unexpected situations. Absent public pressure, there's really almost no incentive at the current scale they operate at for autonomous vehicles to pull over to not block traffic. If all vehicles on the road were autonomous, then yeah maybe they would have other incentives because the vehicles stuck behind them would be their own that are losing time, and time is money. Yet as it is now, they're far more incentivized to just stop there and let everyone else deal with the fallout of a car stopped in the middle of the road and the public pressure of that is what sort of gives them incentive to try to balance public scrutiny and their own liabilities.
Although it does matter, I think you might be over-estimating the role of public pressure. The company and the regulators interact directly, and regulators make their own judgements based on that.
We can see it in this case where regulators approved Waymo’s expansion despite a lot of public sentiment against driverless cars by a public that often doesn’t distinguish between companies.
It would be interesting to know more about how they handle pull overs. Did Cruise change their algorithm? Did Cruise do it differently than Waymo? I don’t think reasoning from incentives is enough to answer these questions.
It's possible I'm overestimating it, but I think the issue in this case is that the regulators approving it are still representing the public, it's just not the local population in SF. So you can't equate regulatory action with public sentiment unless you factor in who is being represented.
The top comment in this thread is talking about people who want autonomous vehicles, I think that is aligned with what a lot of people want and the public in general is likely more tolerant of their faults when the vehicles aren't on their streets. From what I've seen reported so far, SF politicians don't want the vehicles there, representing the public sentiment in their city, but they have no authority to stop them because CPUC which is a state regulatory agency has the authority. The state regulatory agency is more representing an entire state of people who much like the top comments in here probably also want autonomous vehicles and aren't personally inconvenienced by them unless they live in SF (obviously as this post mentions LA is a city Waymo is expanding into, but til this point the residents of LA haven't really had to experience much of disruption from autonomous vehicles).
The problem with knowing how they handle things is that they control all the information about what incidents occur or nearly occur with these vehicles, and only report what they want to report or are required to report, but don't otherwise engage in much transparency to allow the public to understand the development of autonomous vehicles. Reasoning from incentive is simply what is left when there's a lack of transparency and inability to access detailed information about how the vehicles otherwise operate.
The representation is indirect and rather fuzzy. I assume regulators consider it their mission to protect the public, but that doesn’t mean they agree with any particular proposal about how to do it, even if it’s popular. And yes, it’s a state agency, not a local one.
Sometimes we do get pretty detailed reports about what’s going on. Also, these services operate in public and other people report on what they do. Every accident makes the news. That’s one way regulatory agencies would find out if the company isn’t reporting things accurately to them (as happened with Cruise).
Yeah, I think that's the real kicker here. Accidents will still happen because no amount of sensors can defy physics, and early adopter technology will inevitably act in ways humans won't as they adapt to what humans expect.
But trying to more or less hide a life threatening event under PR because shareholders or whatever... it really shows where their priorities lie and why we need to regulate/sanction this behavior now rather than later.
Tbh Uber is still much better than Taxis. A few of them have tried to make apps in the meantime, and god it shows how hard it can be to make apps.
Kinda dumb if no one's thought of building a taxi app and licensing it all over the place.
When there's something that seems simple that would be a great idea that no one's implemented, it's probably because it's not so simple.
Dumb probably was a poor choice of words, since I didn't mean it would be easy but just that it's surprising if no one with the expertise hasn't tried it.
Obviously it would need initial seed money and a lot of legwork since you wouldn't be marketing it directly to consumers. And it's not like the traditional taxi industry is doing that well, plus Silicon Valley probably favors the Uber model, as you can directly exploit the workers, when here there would be a person in the middle.
Still surprising that someone with the means hasn't done it (well).