Hacker News Reader

Comment by ajross | Hacker News Reader

Oh, for goodness sake. Just Google "Tesla accident rate" and you'll find plenty of yammering on both sides of the argument. But read the numbers therein and you'll find that (1) they're noisy, because it's just one car model and (2) they're clearly in line with what you see with other car statistics.

There are plenty of cars that have been detectable death traps. Teslas aren't among them. You're just spooked by the scary computer behind the wheel.

View on HN

You’re conflating two systems. Tesla rolled out Collision Avoidance Assist and Autopilot around the same time. Collision avoidance is undoubtably a great feature that’s contributed to a decline in accident rates. That’s primarily what helps you in cases where you “fail to remain alert.”

But here I am asking about the other feature, autopilot, i.e. autosteering. The TOS requires you to remain fully alert when using the feature. So even if it were pitched as a safety benefit that keeps you from lane drifting when you’re not paying attention — you’d be violating the TOS.

So what is the benefit that doesn’t violate TOS? Clearly the implication is — and most Tesla users agree — it lowers the cognitive load while driving. But this would seem to undermine Tesla’s defense that drivers must remain vigilant and attentive at all times when using it.

Filtering out the noise, you’re not willing or able to cite, and I don’t blame you, since you’d probably try to start with the discredited Tesla numbers. Then I’d post the reasons why they’re crap, and you’d grope for some more apples and oranges numbers. I like how you’ve just skipped past all of that and gone straight to the personal attacks, it saves a lot of time. Since you’ve made your claim without a hint of support, here’s my equally unsupported counter-claim: Tesla autopilot is much less safe than a human who wouldn’t accelerate into a cop car, fire truck, or gaurd rail unless dead, or drunk.

Reply on news.ycombinator.com