Resultado de búsqueda
Hace 2 días · From skimming the paper, it appears that the authors have missed that the central AI Safety measure being argued for is pausing/halting. The "Climb to Armageddon" example very poorly matches the safety proposals by Eliezer Yudkowsky.
Hace 3 días · Last visit was: Fri May 31, 2024 11:19 am. It is currently Fri May 31, 2024 11:19 am
Hace 4 días · Eliezer Yudkowsky and his pals over at lesswrong.com have been talking about those threats for years. In fact, Mr. Schmidt goes on in this video to describe one hypothetical: a nuclear war games scenario, and in a recent talk at Harvard he described another where a recursive agent sucks try electricity from a hospital after learning it needs electricity to complete its task successfully.
Hace 2 días · Brain Science is hosted by Virginia "Ginger" Campbell, MD (Podcast Hall of Fame 2022). She is an experienced physician with a passion for exploring how recent discoveries in neuroscience are revealing how our brains make us who we are. This podcast is for non-scientists, scientists, and everyone in between.
Hace 2 días · Before January 1 2025. Yes. No. Show 1 more answer. Resolves based on when Ilya confirms where he will work next as a full-time employee, as defined in this market: /Joshua/where-will-ilya-sutskever-work-next. I will resolve options that have passed to NO as they pass. Once we know, all others resolve YES.
Hace 4 días · (Eliezer Yudkowsky has sometimes been accused of wanting to bomb data centers, but he supports international regulations backed by military force - his model is things like Israel bombing Iraq’s nuclear program in the context of global norms limiting nuclear proliferation - not lone wolves.
Hace 5 días · P(A|B) = [P(A)*P(B|A)]/P(B), all the rest is commentary. Click to read Astral Codex Ten, by Scott Alexander, a Substack publication.
