[shuwusxuu2he]

Prosecutors, even high profile/positioned ones, have a finite amount of resources. At the end of the day it might be romantic to go after a megacorp but a prosecutor's job is to persue justice against criminals, you might get a few high profile executive types and justice for one family (if you're the type to consider beyond the defendant) but with those same resources and time you could probably lock up at least twenty violent, in the real way, deadbeats. I'm not saying that's the way it should be, but I think it's worth acknowledging because it's why the problem is hard to solve. Everyone wants to see big fish getting hard time, but reality looks more like Harvey Dent's speech to Commissioner Gordon and that's for the good prosecutors.

[dodobirdlord]

But moreover, it's totally unreasonable that anyone would have gone to jail as a result of this, except again, maybe the driver. The driver was literally being paid to monitor the vehicle. How is an argument that high profile executive types were engaging in criminal negligence going to fly, when the bar for legal driving is already that 1 person is in the car nominally paying attention?

[Super_Jambo]

Because the driver was given an impossible task in order to save money for the company. They were required to both fill in information on a tablet and monitor the car. It's not even clear that monitoring a semi-self driving car is even possible. Constant vigilance for the 2% of edge cases where it might crash is harder than just driving. Uber had previously used two people to do this task but had switched to one person.

Executives must be prosecuted when their money saving results in deaths or else there is no incentive for them not to make this trade off.

[omarhaneef]

Not taking sides here but want to point out one thing:

When I took the Uber self driving cab in Pittsburgh (randomly called on my Uber), there were two people in the car.

One person was a driver, and the other person was in the passenger seat with a laptop. They were both very polite. The person with the laptop was very focused and watched the sensors take in information.

The driver -- at least in my case -- was very cautious and almost never let the car do anything by itself. He was almost driving it the whole time.

This is right after Uber brought over the team from Carnegie Mellon (so 2015 ish?).

[SirSourdough]

Yeah, but out of all the self-driving test miles Uber is doing, how many are in the self-driving cabs vs non-public road testing?

I would expect them to be very careful with a passenger in the car scrutinizing but much less so in a test car. If a high enough percentage of the millions of miles these companies are doing are tests without passengers, there could be a lot of road miles happening with a single overburdened driver that the public isn't seeing much of.

[roenxi]

There are a bunch of incentives for them to get this right. Moreover there might be a reasonable argument that they are having a net positive impact even if their car was slightly less safe than most drivers - they are going to save people probably single digit % points of their life staring at a road.

If you think their car isn't safe; don't buy one and don't be driven in one. It isn't reasonable to say they can't have the occasional accident though.

[agustif]

The person who got hit probably didn't buy one or was driven by one. just hitted.

[xnyan]

Thanks, I'll let Uber know I don't consent as my insides deform from impact of the hood striking my torso.