3 ms·
100,000 regular cars didn't make the decision to kill their occupants though. Drunk drivers kill people so we try to prevent drunk driving, If cars are occasio
by bogantech 4y ago
100,000 regular cars didn't make the decision to kill their occupants though.
Drunk drivers kill people so we try to prevent drunk driving, If cars are occasionally killing people due to a design flaw they should be investigated.
Don't worry though - Tesla will say that autopilot deactivated 10ms before impact and that it's all the drivers fault and people will go back to ignoring it
- skellington 4y agoIf drivers make deadly mistakes at X rate, and software makes deadly mistakes at Y rate where Y < X, should we recall the software? What if Y is substantially less than X but still not zero?
- kube-system 4y agoThe complication with that is that software makes different mistakes than drivers do. And sometimes, software refuses to drive, in which case, their accident rates are subject to selection bias. For cars which can be operated by software or a driver, you’d want to compare for an exact equivalent scenario.
- treefry 4y agoThe problem is Y is more uniform but X is not. Good drivers make significantly fewer mistakes than the average. For these people, using software potentially increases risk by order of magnitudes.
- speedgoose 4y agoThe autopilot may deactivate itself in such a situation, it tends to not like uncertainty, but as far as I know everyone will still consider that a crash under autopilot.