The company left out some key details regarding the incident involving one of its robotaxis and a pedestrian.
The company left out some key details regarding the incident involving one of its robotaxis and a pedestrian.
On October 2, 2023, a woman was run over and pinned to the ground by a Cruise robotaxi. Given the recent string of very public malfunctions the robotaxis have been experiencing in San Francisco, it was only a matter of time until a pedestrian was hurt by the self-driving cars. New reports, though, suggest that Cruise held back one of the most horrifying pieces of information: that the woman was dragged 20 feet by the robotaxi after being pushed into its path.
The LA Times reports:
A car with a human behind the wheel hit a woman who was crossing the street against a red light at the intersection of 5th and Market Streets. The pedestrian slid over the hood and into the path of a Cruise robotaxi, with no human driver. She was pinned under the car, and was taken to a hospital.
But this is what Cruise left out:
What Cruise did not say, and what the DMV revealed Tuesday, is that after sitting still for an unspecified period of time, the robotaxi began moving forward at about 7 mph, dragging the woman with it for 20 feet.
I remember someone here chiding others for critizing Cruise. They were talking up the "fact" that the car stopped and let emergency services dictate what to do instead of risk harming her further. It was GOOD that the car stopped on her.
No matter what happens, anything or anywhere. There will always be people defending it. I wonder if that person is glad to hear the car heroically dragged her out of traffic?
I think a lot of people forget that machines and AI are not infallible gods. Like a few dings from some highly charged particles from space and next thing you know they flip a 1 and 0 somewhere in their code and run that instead.
Tell that self driving car "experts. Because there is a silicon chip controlling they think it's automatically better than a human. Said experts have never heard of random failures (that's besides bugs).
On the other hand, there's people who will condemn the very concept of self-driving cars because of the kind of event that happens every day with regular ol' human-piloted cars.
This is a serious incident, it should be thoroughly investigated, regulations on safety and reporting should be seriously considered. But don't strangle the baby in the crib: self-driving cars have the potential to be much safer than human-driven cars (arguably, they already are). When there's stories about Cruise taxis stopping in an intersection and the response is an overwhelming flood of "ban all self-driving cars!", it causes proponents to get overly defensive.
When there’s stories about Cruise taxis stopping in an intersection and the response is an overwhelming flood of “ban all self-driving cars!”, it causes proponents to get overly defensive.
But the problem is those events are happening. People are wanting to strangle that baby because the kid is causing harm in the real world. The cars DO need to be taken off the streets, but that doesn't mean they should be outlawed. However, they absolutely are not fit for public service. As someone who has "driven" a Tesla while it was "Full Self Driving", and I gotta tell ya that it just can't handle the rules of the road. It's not better than a human. We can adapt to new situations on the fly, but a self driving car can only operate in certain parameters.
While I sympathize that there is nothing you can do about this, the real fault lies with the companies putting vehicles on the road before they are ready. Without a doubt, someday they will be. But as corporations push this out the door in the rush to be the first and the new name brand, they do so at the public expense. It's the root problem for the cycle of backlash, resentment and defensiveness.
That's why we have grade-separated rails, and stations with barriers and doors to the train. Trains also have a fraction of the deaths per passenger mile vs cars.
Public transit has a variety of benefits. For one thing, the natural enemy of the driver is literally other drivers. Cars are very space intensive, so car-centric cities tend to sprawl.
Public transit supports walkable/bikeable density, because it does better with a good walkshed around stations and also has really good passenger throughput. That's good for people's health - people living in walkable areas are on average less sedentary, and have lower rates of obesity and diabetes. It also tends to be good for the creation of third spaces, and seems to be good for social engagement on average.
The reason to oppose self-driving cars is really the same reason to oppose car-centric infrastructure broadly: the alternatives are way better.
The pedestrian slid over the hood and into the path of a Cruise robotaxi, with no human driver. She was pinned under the car, and was taken to a hospital.
My vision of self driving cars was of an integrated system where all the parts weave together to create a safer and faster environment. But self driving cars are just not able to deal with the edge cases that will pop up. Even that would be okay, but GM tried to cover up this horrific accident. That inspires the opposite of trust. I gotta wonder how many other incidents have been covered up. GM is a company with limited resources. Alphabet, the parent company of Waymo, has a virtually infinite budget. How many incidents have they hidden from the public eye?
Anyone with even a hobby level of coding knowledge knows it’s solving the edge cases that’s the real issue with resolving software problems. In this case I wouldn’t be surprised if automation reaches similar or lesser levels of traffic incidents, but the real shitty part is gonna be how much harder it is to get justice from a large corporate entity owning these robotaxi fleets versus nailing the little guy driving his Uber/taxi.
I don’t fully disagree but they have clearly tried to jump to market without fully investigating the problems. Things like this need serious regulation and that doesn’t exist right now.
These cars should be monitored by human beings until their AI evolves enough to be actually more secure than human professional by-the-law pilots. If a human was monitoring the car, they probably could have stopped it immediately, or even hold it before it starts dragging that poor woman.
Only if these cars can do the same or better than the human overseeing their activity, these cars will be safe enough to be offering a public service. Also, as shameful as it could be, this incident must get the most publicity because other competitors should test their AI against this specific situation as soon as possible.
I mean, humans are perfectly capable of not knowing someone is trapped under the car and doing something like this. It's awful, but it happens pretty regularly. Pulling over to the side after an accident is a pretty heavily ingrained thing.
In this case, it's not the technology that scares me, but the company developing it not being honest.
The car could have a safety record vastly better than any human, but if the company making it isn't transparent about incidents it entirely undermines our ability to trust that safety record.
I worked in this industry as a safety driver and various other positions for over 6 years in AZ, first with a company that also made national headlines, and then with the company that has connections with a certain search engine. From the inside it's easy to see these outcomes happening more and more frequently, these companies are concerned with getting as many driverless miles as they can because that's where their data comes from, the data is where the money comes from and that's all that matters. Drivers are typically subcontracted, forced to work long and stressful hours with few breaks. Safety is emphasized, but even dealing with fatigue in the appropriate ways can lead to disciplinary action if you're fatigued too often, so it goes unreported. I left the company specifically for safety concerns and despite making double there what I make now I won't be going back.