News for hackers. Stuff that matters.
Re:Statement on AI Risk by lambertsimnel on Tuesday May 30, 2023 @10:32AM (#36125310)
I recommend Robert Miles's YouTube channel, and his "Intro to AI Safety, Remastered" is a good place to start: https://www.youtube.com/watch?v=pYXy-A4siMw

I find it worryingly plausible, not that AGI is certain to go badly, or even that if it does go badly it will necessarily lead to human extinction, but as Robert Miles says, "If you have a sufficiently powerful agent and you manage to come up with a really good objective function, which covers the top 20 things that humans value, the 21st thing that humans value is probably gone forever" (about 12:40 into the video).

There are no comments on this comment yet.