3 ms·
That's what they'll use for autonomous killer robots, too. Not hard evidence that someone is a real threat, but "predictive assessment". It's basically what th
by mtgx 11y ago
That's what they'll use for autonomous killer robots, too. Not hard evidence that someone is a real threat, but "predictive assessment".
It's basically what they do with drones strikes, too. The only difference is that someone has to sign-off on it (still rather important check, but it has likely become another rubber-stamp action by now, too) and then someone else has to push the button (but has not play in deciding whether the target is innocent).
Once again, data mining for stopping terrorist attacks doesn't work (and it creates orders of magnitude more false positives):
https://www.schneier.com/essays/archives/2005/03/why_data_mining_wont.html https://www.schneier.com/essays/archives/2005/03/why_data_mi...