Mobileye Says Tesla Was Dropped Because of Safety Concerns

An anonymous reader writes: On Wednesday, Mobileye revealed that it ended its relationship with Tesla because "it was pushing the envelope in terms of safety." Mobileye's CTO and co-founder Amnon Shashua told Reuters that the electric vehicle maker was using his company's machine vision sensor system in applications for which it had not been designed. "No matter how you spin it, (Autopilot) is not designed for that. It is a driver assistance system and not a driverless system," Shashua said. In a statement to Reuters, Tesla said that it has "continuously educated customers on the use of the features, reminding them that they're responsible to keep their hands on the wheel and remain alert and present when using Autopilot" and that the system has never been described as autonomous or self-driving. (This statement appears to be at odds with statements made by Musk at shareholder meetings.) It is also emerging that the crash which cost Joshua Brown his life in May of this year was unlikely to have been the first such fatal crash involving Tesla's Autopilot. In January of this year in China, a Tesla ploughed into the back of a stationary truck at speed, killing the driver.

Comments

Unreasonable

[AmiMoJo]

Asking customers to remain alert while the car drives itself for hours on end is unreasonable. Psychologists know that, NASA warned them about it... Human beings simple can't concentrate for that amount of time with nothing to do.

Re:Unreasonable

[stanjo74]

Survivor bias. You were not alert, just lucky.

Re:Unreasonable

[ShanghaiBill]

Exactly. If you call something autopilot, then people expect it to be an... autopilot.

... unless they are actually a Tesla owner. I use Autopilot, and Tesla repeatedly and emphatically makes the capabilities of the system and the responsibility of the driver very clear.

Re:Unreasonable

[I4ko]

His point exactly - YOU drove, you weren't driven.

Re:Unreasonable

[Dare nMc]

>Asking customers to remain alert while the car drives itself for hours on end is unreasonable.

It is also be unthinkable to have your "backup" to evaluate the performance of the autopilot watching only the output. I have hundreds of hours logged in autonomous vehicles, but I would review the data, see all the diagnostics logged, all of the GPS signal lost, or drifted, etc for the week. I thus have never completely trusted them. All of the operators, even when told by engineers of running a beta release with big untested changes would spend all of their time working on their phones. Without knowing when every backup and sensors have failed to read something wrong. You cannot evaluate the maturity just off of, well it stopped the other 5 times someone stepped in front of the vehicle, why would I have to worry about walking in front of them. If you don't know 15 times in the last mile the cameras failed to maintain the road edge monitoring, and 20 times during that same period that sensor was the only thing that kept you on the road. Only dad those 2 events overlapped, which they eventually will, would the failures actually show up in the output.

Well...

[johnsmithperson123]

Tesla has about 2 fatalities per 100 million miles. South Carolina, the worst state in the US for accidents, has 1.65 accidents/100M. Massachusetts has .57. Clearly, self driving cars have a long way to go.

Re:Well...

[mbeckman]

Statistically insignificant. Tesla stats will only matter when tens of thousands, if not millions, of trips have been made under autopilot. Then compare accident rates. If Tesla turns out to be safer, it won't be because of AI, because we have no idea how humans drive in the first place. It will be because of image processing and predictive algorithms, combined with pre-ordained decision trees. And there may well be major unforeseen consequences, such as cascading failures and catastrophic feedback loop interactions between vehicles.

Mobileye understands lit. Musk doesn't.

[HBI]

Expecting Tesla to survive the avalanche of product liability suits that are coming is crazy. Musk appears oblivious to the problem. This is not a PR issue. There are numerous chinks in Tesla's armor that will be pried open and exploited by plaintiff lawyers. The company is toast. Mobileye is just trying to save itself and preserve relationships with other vendors.

As for that idiot Hotz...we can go visit him in some slum apartment in a few years. Bring a 12 pack and you can listen to him complain about how the system is rigged.

Re:Mobileye understands lit. Musk doesn't.

[BostonPilot]

Expecting Tesla to survive the avalanche of product liability suits that are coming is crazy. Musk appears oblivious to the problem. This is not a PR issue. There are numerous chinks in Tesla's armor that will be pried open and exploited by plaintiff lawyers. The company is toast.

I'm glad you mentioned this. Just this weekend I telling (another pilot) that I don't understand the strategy. The goal of Tesla was to bring electric vehicles to the masses. How are they going to do that when they get sued into oblivion? A conservative approach would have been to offer assist technologies similar to what their competition (other luxury brands) was offering. Instead, Elon has acted like it's Autopilot that's selling Tesla cars. I think people like Autopilot, but would buy the car if it had a much less aggressive auto-drive system because the real value is in the electrification of the car, not the autopilot system.

It's not all that dissimilar to his falcon wing door misstep, except that falcon wing doors did not present an ongoing risk of expensive lawsuits.

So far the accidents have been such that the Tesla driver was the one who got hurt. What happens when a Tesla hits another car and kills everybody inside? How is Tesla going to avoid the liability? Yeah, sure, the driver should have been paying attention, but at least in the US Tesla will still get named in the lawsuit, and when they lose guess who is going to have to pony up the majority of the settlement? Hint: it won't be the driver.

The good news is that Elon may have already jump-started the electric car industry and even if Tesla gets sued out of existence we may have enough momentum for the other car companies to keep moving forward.

Re: I'm short TSLA

[lxs]

Shorting a Tesla is generally a bad idea. 60-85kWH batteries tend to react vigorously to that sort of treatment.

Re:Gives new meaning to computer crash

[Rei]

Or not. I love the weasely " In January of this year in China, a Tesla ploughed into the back of a stationary truck at speed, killing the driver. Should that incident prove to be related to Autopilot...". Well, yes, in the same way that if a train were to crash tomorrow you could write "Should that incident be related to the Galaxy 7..." without any evidence that it was involved at all.

Here's a Google Transmangle of the original article in Chinese:

https://translate.google.com/t...

Basically, the evidence that autopilot was in use was... um... his dad thinks it must have been because his son is a good driver and wouldn't have hit that truck. And he wants to "prove" it by... showing that the car's speed wasn't changing.

Whether the autopilot was on or off in a given situation is logged and easy to recover. Any reporter who suggests that an incident was "due to autopilot" without at first finding out whether the system was even on is being grossly irresponsible.