Because there is no evidence of such a third person.

If there was a third person, why not a fourth, and a fifth, and a sixth? Or a dozen, or two? It would also be far from the first time that a troupe of clowns fit in a too-small car.

And what evidence usually exists of a third person in a non-Tesla missing-driver crash?

Look, all I'm saying is that if this was literally any other car, we would all see this evidence and say "Yup, looks like a drunk driver fled the scene of an accident". It would be an obvious conclusion. We wouldn't be scratching our heads saying "Well, gee whiz, maybe there's some other explanation we haven't thought of".

This is not the first time police have found a crashed car with no driver. The only difference this time is that it's a Tesla.

The difference is that there were two people in the car, one of which was the driver (and owner) of the car.

But you are right to suggest that the police are themselves jumping to conclusions. Just because the accident happened in a tesla, doesn't mean that self-driving was somehow involved.

I don't disagree about that. I disagree about the necessity of assuming a third person, when the two present suffice to have caused the accident and when there is nothing to indicate that a third person was involved.

Finally, assuming a third person introduces a new question: who was it?

Do you know?

The evidence of the third person is that otherwise you've got to explain how the people in the car tricked the autopilot into driving at crashable speeds in a tiny cul-de-sac, where normally it wouldn't even engage, even apart from tricking it that there was a driver in the seat.

At this point, it's an Occam's Razor question, though different people will argue about which was simplest.

If there had been a third person who fled the scene, though, you would expect any physical evidence to have been left behind after that fire.

>> The evidence of the third person is that otherwise you've got to explain how the people in the car tricked the autopilot into driving at crashable speeds in a tiny cul-de-sac, where normally it wouldn't even engage, even apart from tricking it that there was a driver in the seat.

But that's not "evidence". That's just an unanswered question.

As I said recently in another comment, if you find yourself at the point where you're thinking "I can't think of a better explanation", that's not the time to say "that must be the answer"; that's the time to think harder until you find a better explanation.

In this case, if you can't find a better explanation than "there must have been a third person that we don't know was there", then it's time to sit down and think harder about how the accident could have really happened.

Because if you start assuming things we have no evidence for, you might as well assume invisible space leprechauns, for all the good it will do.

> you've got to explain how the people in the car tricked the autopilot into driving at crashable speeds

Or the autopilot did that by itself. I don't see why we need to assume any human trickery.