4 ms·
What jimrandomh is talking about is not just whether megacorps use AI for good or bad, but what happens if an AI starts doing damage outside the control of anyb
by dwohnitmok 4y ago
What jimrandomh is talking about is not just whether megacorps use AI for good or bad, but what happens if an AI starts doing damage outside the control of anybody. That could happen regardless of democratization if advancement occurs too quickly without thinking about AI safety.
- TigeriusKirk 4y agoWhat are the dangers?
- jimrandomh 4y agoExactly. To flesh out the details a bit, an example of how this could go is: https://www.lesswrong.com/posts/kpPnReyBC54KESiSn/optimality-is-the-tiger-and-agents-are-its-teeth https://www.lesswrong.com/posts/kpPnReyBC54KESiSn/optimality... .
- cdomigan 4y agoThis seems terrifyingly plausible