Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
For the particular accident, then the manufacturer would be named defendant and a jury would proportion the damages accordingly. I suppose if there were a string of accidents caused by such a defect, then a class action against the manuf. would be possible, but that’s outside the scope of the discussion.

Again remember many here are discussing who would bear ultimate responsibility at the end of trial - after all fact finding and such. I’m not arguing that. I’m saying that our current system is perfectly capable of sussing out all these issues and assigning responsibility without any new regulations. There is no “liability problem” in our court system that needs to be worked out for autonomous cars to be a thing.

But it does require all parties to present. If they were to pass some legislation that said, as it appears some are suggesting, that in an accident involving an automobile operating under an SAE level 3+ autonomous driving system, the owner/operator bears no liability and only the manufacturer may be held responsible, then I think that would be a real problem.
You can sue just about anyone but I have no idea why you would sue the owner for a design defect when the manufacturer is the one with all the money.

Why would it be a "real problem" if the manufacturer were solely liable for design defects?
 
  • Like
Reactions: DanCar
It will be interesting to see how this ultimately plays out but I am of the opinion that as far as Tesla is concerned their legal team will find ways to divert and deflect liability. GM will likely do the same. I can’t see the shareholders agreeing to additional risk when they don’t have to.
All of this is super fascinating to watch and discuss, one of the more interesting things that I've spent any time researching in the last few years
 
Never said that. I said good luck defending yourself against Mercedes and their deep pockets. But then again there was Volkswagen and diesel gate.
You said the Mercedes would fabricate log data showing that driver did not take over in time and present that in court. That is criminal.
Also, this prediction is absurd as the system just stops in the middle of the road if the driver doesn't take over. There really isn't a possible scenario where your proposed scheme would work.
 
  • Like
Reactions: nvx1977 and DanCar
But then again there was Volkswagen and diesel gate.


In cases where the vehicle crashed prior to requiring a human takeover then Mercedes is 100% at fault, per their own admission. The depth of their pockets is irrelevant.

There's nothing to "defend" against for the owner of the car- who was responsible for the DDT when the accident happened will be clearly visible in the legally required data recorder logs.

Your claim no MFG would ever claim responsibility is factually wrong. If you wanna keep digging a deeper hole on this maybe ask The Boring Company for a consult.
 
  • Like
Reactions: AlanSubie4Life
In cases where the vehicle crashed prior to requiring a human takeover then Mercedes is 100% at fault, per their own admission. The depth of their pockets is irrelevant.

There's nothing to "defend" against for the owner of the car- who was responsible for the DDT when the accident happened will be clearly visible in the legally required data recorder logs.

Your claim no MFG would ever claim responsibility is factually wrong. If you wanna keep digging a deeper hole on this maybe ask The Boring Company for a consult.
You don’t have to agree with me. I stand by my opinion that this is all PR by the companies. When reality sets in they will put it all on the driver. Let’s wait and see as this plays out in the real world.
 
You don’t have to agree with me. I stand by my opinion that this is all PR by the companies. When reality sets in they will put it all on the driver. Let’s wait and see as this plays out in the real world.
Yeah, that would be great PR.
It's far more likely that they'll realize that self-driving is a lot harder than they thought, disable the L3 system and issue refunds.
 
Despite the fact that Mercedes' system is sold as level 3, I think they're banking on three facts:

1. It won't be equipped on too many vehicles or operate in many scenarios, so the total number of miles driven on L3 it will be relatively low
2. Given the limited ODD, most drivers will find it too annoying to bother activating in the first place. Or they'll activate it a few times as a novelty when the car is new, but not rely on it often.
3. Being concerned for their cars and lives, most drivers will end up treating the system as if it were level 2 anyway
 
That doesn't really make sense because the manufacture (or fleet operator) is responsible for the accident. Now he could be talking about L2 FSD style mode where the driver is still responsible.
Yea, you know I was thinking the same thing. Then again, nothing with insurance companies makes much sense to me. What he was saying is that they evaluate whether or not being in a self-driving mode actually reduces accidents and injuries compared to manual driving. And if it does, they might give an incentive to waive part or all of a deductible if any accident occurs while in self driving. Their biggest focus is on medical injuries since that is their bigger part of accident cost and typically also involves attorneys. It seems that so far, they are not convinced this is the case, yet continue to review rates and statistics from time to time as cars advance and more car companies include things like "Super Cruise" whatever that is. We shall see if policy changes ever occur. I'm not convinced yet either.
 
  • Like
Reactions: S4WRXTTCS
You can sue just about anyone but I have no idea why you would sue the owner for a design defect when the manufacturer is the one with all the money.

Why would it be a "real problem" if the manufacturer were solely liable for design defects?
I think we are arguing two different things. It's not a problem for a manufacturer to be solely liable for design defects - I never said otherwise. And, in fact, the manufacturer would be solely liable for design defects, assuming there is liability for the design defect, which varies from state to state. What I said was it would be a problem if operators/owners could depend on legislation that shielded them from liability (and thus being named in a suit for a particular accident) if their automobile was operating under an SAE L3+ autonomous system.

For example, how do you know that a design defect was responsible for (i.e., was the proximate cause of) a particular accident without a fact finder (e.g., a jury) determining such? And even then there may be liability for the accident on the owner/operator. What if the owner/operator was aware of the design defect and operated the car in the L3+ mode anyway? What if the owner were part of a beta test program and complained vehemently about the safety of the system on a social media website right up to its general release (at the end of the year) and then immediately began operating it in an unsupervised L3+ mode based solely on the assurances from tweets from a rather unreliable CEO of the automobile manufacturer that it was completely safe? I think I would argue that owner/operator had some liability for the accident (yes, if there were money to pay damages there).
 
Last edited:
What I said was it would be a problem if operators/owners could depend on legislation that shielded them from liability (and thus being named in a suit for a particular accident) if their automobile was operating under an SAE L3+ autonomous system.

Except if you're in L3 self driving mode, you are not an operator- you are a passenger

By definition.

I'm unaware of any passenger being sued for liability in an accident outside of perhaps outliers like "Passenger grabbed and yanked the wheel from driver" or something.


What if the owner were part of a beta test program and complained vehemently about the safety of the system on a social media website right up to its general release (at the end of the year) and then immediately began operating it in an unsupervised L3+ mode based solely on the assurances from tweets from a rather unreliable CEO of the automobile manufacturer that it was completely safe? I think I would argue that owner/operator had some liability for the accident (yes, if there were money to pay damages there).

This question doesn't even make sense.

FSDBeta is an L2 system that ALWAYS requires a human driver to perform part of the DDT. Tesla repeatedly tells you this in the release notes, in the thing you have to agree to to turn it on, and in every official government document they've discussed the topic in.

You can not "choose" to operate a system at a higher level than the designer says it's capable of-- any more than putting on a blindfold in a totally dumb car is the driver "choosing" to pretend the car can see for him.

The driver in the case of a system the maker says is L2, but you pretend is L3, the accident is on the driver. Who is a human. Because that's how L2 works by definition.
 
I think we are arguing two different things. It's not a problem for a manufacturer to be solely liable for design defects - I never said otherwise. And, in fact, the manufacturer would be solely liable for design defects, assuming there is liability for the design defect, which varies from state to state. What I said was it would be a problem if operators/owners could depend on legislation that shielded them from liability (and thus being named in a suit for a particular accident) if their automobile was operating under an SAE L3+ autonomous system.

For example, how do you know that a design defect was responsible for (i.e., was the proximate cause of) a particular accident without a fact finder (e.g., a jury) determining such? And even then there may be liability for the accident on the owner/operator. What if the owner/operator was aware of the design defect and operated the car in the L3+ mode anyway? What if the owner were part of a beta test program and complained vehemently about the safety of the system on a social media website right up to its general release (at the end of the year) and then immediately began operating it in an unsupervised L3+ mode based solely on the assurances from tweets from a rather unreliable CEO of the automobile manufacturer that it was completely safe? I think I would argue that owner/operator had some liability for the accident (yes, if there were money to pay damages there).
Well Tesla is the only company having customers test beta L3+ software so I doubt that will be the norm. I guess I'm making a smart decision to wait until FSD is out of beta!
I'm just having trouble thinking of a collision where the car is at fault that would not qualify as a design defect (in states that have adopted the SAE taxonomy in their autonomous vehicle legislation).
It certainly doesn't seem like a problem to me if passengers in a car being operated in accordance with the manufacturers instructions are not liable if the car crashes. It seems logical.
 
It also might be that L3+ systems are designed to never crash into something - their "prime directive". To that end, the car will avoid, at all cost, being "at fault" for an accident. That may mean it slams on the brakes and causes an accident, such as another car rear-ending it, and possibly a chain reaction after that. But in the end the car will not be "at fault" and therefore not liable. 🤔
 
Except if you're in L3 self driving mode, you are not an operator- you are a passenger

By definition.

I'm unaware of any passenger being sued for liability in an accident outside of perhaps outliers like "Passenger grabbed and yanked the wheel from driver" or something.




This question doesn't even make sense.

FSDBeta is an L2 system that ALWAYS requires a human driver to perform part of the DDT. Tesla repeatedly tells you this in the release notes, in the thing you have to agree to to turn it on, and in every official government document they've discussed the topic in.

You can not "choose" to operate a system at a higher level than the designer says it's capable of-- any more than putting on a blindfold in a totally dumb car is the driver "choosing" to pretend the car can see for him.

The driver in the case of a system the maker says is L2, but you pretend is L3, the accident is on the driver. Who is a human. Because that's how L2 works by definition.

You are technically right, but i think the whole discussion about liability in the last few pages is about what will insurance companies and victims do when an actual OEM releases a car with L3. Today there are no L3 cars in production so we don't actually know...just that theoretically there is no liability for owner when in L3/L4 self driving mode.
 
  • Like
Reactions: Goose66
You are technically right

Cliche_technically_correct.jpg




, but i think the whole discussion about liability in the last few pages is about what will insurance companies and victims do when an actual OEM releases a car with L3. Today there are no L3 cars in production so we don't actually know...just that theoretically there is no liability for owner when in L3/L4 self driving mode.

They're welcome to sue the person who was a passenger in the car

Just as they're welcome to sue passengers in accident causing cars today.

But they can expect the same, disappointing for them, results when doing so.

(that said- this is specific to Mercedes who stated they would take liability while the car was doing the DDT....remains to be seen if anyone else does, but it does debunk the earlier claim we'd "never" see an OEM doing that)
 
2. Given the limited ODD, most drivers will find it too annoying to bother activating in the first place. Or they'll activate it a few times as a novelty when the car is new, but not rely on it often.
I think you're on to something here. During one review the journalist couldn't keep it engaged. Traffic wasn't slow enough, then when it did get slow it was near a bridge or something that wasn't an approved area, then when it got past that traffic sped up again. Then if I recall traffic slowed again but it was a construction zone, which requires handover back to the driver. And so on....
 
most drivers will find it too annoying to bother activating in the first place. Or they'll activate it a few times as a novelty when the car is new, but not rely on it often.
Lol, sounds like something else we know about, in its current state.

It does seem like conditions for use of Mercedes system would be really rare. I actively avoid ever driving in such traffic, and it’s so common for things to be slow, then speed up, etc.

Still, it’s a start. Baby steps I guess.