Hacker News new | ask | show | jobs
by lkesteloot 3381 days ago
But if those deaths were to occur tomorrow, absolutely the ends would justify the means. There's no judge in America who would kill 10k people tomorrow just to do justice to Uber. But because those are in the future (and as you say, hard to prove now), it's not even considered. But they're real deaths of real people.

There's going to be a year when self-driving cars become ubiquitous and deaths from cars drop to almost zero. The previous year, many thousands will die in cars. Those deaths are avoidable by pushing this agenda forward, and preventing Uber from moving forward hurts the agenda.

4 comments

> hard to prove now ... But they're real deaths of real people.

That's a contradiction in my books. I don't want to get into a metaphysical discussion of what "real" is, but I challenge you to say that things that may not happen are real. Example: Landing humans on Mars is not real.

> There's going to be a year when self-driving cars become ubiquitous and deaths from cars drop to almost zero.

Yes. I do agree with this.

> and preventing Uber from moving forward hurts the agenda.

This is not fact. I could argue that the greatest barrier to self-driving cars reducing vehicle deaths to 0 is legislation. I could then argue that the most harmful thing to pass that legislation is early, highly publicized failures of that technology. Finally, I would argue that of all the tech companies in a position to implement said technology, Uber is the MOST likely to push it too fast and cause a highly public, negative event (given their current cloud of doom). Therefore, I conclude that the most important thing to save lives is prevent Uber from implementing that technology.

But it's all moot because none of that is real. Just like your assertion.

> Uber is the MOST likely to push it too fast

Sure, I think that's a plausible scenario. It's also plausible that Uber will succeed (they've succeeded a lot!) and this will light a fire under everyone's butt and create amazing competition, pushing all car makers into this space earlier than they would have otherwise.

Of course we don't know, but that doesn't mean we shouldn't guess. My scenario above seems most likely to me, and therefore stopping Uber lets car makers rest of their laurels a bit longer.

And by the way I would take this further: I'd like to government to provide immunity to self-driving car manufacturers for the next few years while they develop their technology, the same way that vaccine makers are immune to health damages from their products.

Remember that we should be willing to kill hundreds of people with self-driving cars just to push the technology forward even one week. (600 people per week die in the U.S. in cars.) Of course I agree with you that this is politically untenable, but it's still what should be done.

> Remember that we should be willing to kill hundreds of people with self-driving cars just to push the technology forward even one week

I will never agree to that. A negligent death is a negligent death. We are not on the same page and I don't feel there is any potential benefit to a continued discussion. Good day.

There's a reason why the legal system judges people on harm they actually cause and not the hypothetical best case scenario from allowing them to continue causing harm.
Privileging theft in this area by not punishing Uber discourages the further investments necessary to develop the technology so that those lives will be saved, setting things back more than any punishment to Uber would.
Allowing Uber to move forward may mean Waymo takes much more time to develop this tech (since they'll be paranoid about any leaks), and meanwhile Uber burns out as many predict. Now allowing Uber to move forward is the killing action.

In the end, we can't know the future, so we use principles that are supposed to lead, by and large and on average, to better societal outcomes. Now, we can discuss specific principles, but reading tea leaves is not useful.

Judges have historically been quite willing to kill people, statistically.

The Supreme Court's decision to block the Medicaid expansion, for reasons, has cost thousands of lives per year for several years.