3 ms·
Autonomous weapons don't decide whether or not to kill someone based on subjective moral criteria. They're dumb. They don't discriminate. They're essentially ju
by onion2k 6y ago
Autonomous weapons don't decide whether or not to kill someone based on subjective moral criteria. They're dumb. They don't discriminate. They're essentially just working on a model that looks like "kill any human within this virtual fence". Building in features that try to be clever is unnecessary and complicated. You just need a sign that says "Don't go in here."
- kjrose 6y agoUntil the day that it messes up and thinks that a certain area outside the fence is within the fence.
- Nasrudith 6y agoThe warnings/fences should be placed past its range limits anyway if done properly given overpenetration and missed shots. Of course friendly fire also should not happen.
- oconnore 6y agoThis feature is ready to ship, our data shows that over 99.2% of hikers correctly interpret our fence boundaries!