3 ms·
And the guy who's already argued for airstrikes on datacenters considers that to be good news? I'd expect the idea of LLMs tending to express a global, triviall
by throwanem 2y ago
And the guy who's already argued for airstrikes on datacenters considers that to be good news? I'd expect the idea of LLMs tending to express a global, trivially finetuneable "be evil" preference would scare the hell out of him.
- thornewolf 2y agoHe is less concerned that people can create an evil AI if they want to and more concerned that no person can keep an AI from being evil even if we tried.
- throwanem 2y agoHe expects the bad guy with an AI to be stopped by a good guy with an AI?
- bdangubic 2y agoworks for gun control :)
- knowaveragejoe 2y agoI hope this is sarcasm because that is hardly a rule!
- mitthrowaway2 2y agoNo, he expects a bad AI to be unstoppable by anybody, including the unwitting guy who runs it.
- DennisP 2y agoNo, he expects the AI to kill us all even if it was built by a good guy. How much this result improves his outlook, we don't know, but he previously put our chance of extinction at over 95%: https://pauseai.info/pdoom https://pauseai.info/pdoom
- imtringued 2y agoThese guys and their black hole harvesting dreams always sound way too optimistic to me. Humanity has a 100% chance of going extinct. Take it or leave it.
- DennisP 2y agoIt'd be nice if it weren't in the next decade though.
- staunton 2y agoI guess the argument there would be that this news makes it sound more plausible people could technically build LLMs which are "actually" "good"...