korrents

On the map

Tap a claim on the ring to put it at the centre.

← We do not control neural networks well enough to guarantee an AI will…

8 connected korrents · 9 moments on record from 17 Feb 2023 to 1 Sept 2026.

Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: We do not control neural networks well enough to guarantee an AI will not harm humans, so we should be careful about what capabilities we give them. We do not control neural networkswell enough to guarantee an AI willnot harm humans, so we should becareful about what capabilities wegive them. Last stated 4 years ago 17 Feb 2023 · 4 years ago AR Armin Ronacher — holds since 2023-02-17 — tap for who they are Same subject: AI takeover is not the thing to worry about: physical constraints bound recursive self-improvement, and humans reliably act once a risk becomes immediate. — tap to centre the map on it AI takeover is not the thingto worry about: physicalconstraints bound recursiveself-improvement, and humans… Last stated 2 years ago 3 Feb 2025 · 2 years ago NL Nathan Lambert — holds since 2025-02-03 — tap for who they are Same subject: Once AI starts to feel powerful rather than mistake-prone, every AI company will become far more paranoid about safety than it is today. — tap to centre the map on it Once AI starts to feelpowerful rather thanmistake-prone, every AIcompany will become far more… Last stated 9 months ago 25 Nov 2025 · 9 months ago IS Ilya Sutskever — holds since 2025-11-25 — tap for who they are Same subject: The likeliest bad path is not a coup but sloppiness: AI does everything verifiable well, research races ahead, and the subtle work of keeping AI safe is what gets done badly. — tap to centre the map on it The likeliest bad path is nota coup but sloppiness: AI doeseverything verifiable well,research races ahead, and the… Last stated 4 weeks ago 11 Aug 2026 · 4 weeks ago RG Ryan Greenblatt — holds since 2026-08-11 — tap for who they are Same subject: Judging an AI by how close it is to a human is the mistake: appreciating what it can actually do is both more accurate and better protection against its dangers. — tap to centre the map on it Judging an AI by how close itis to a human is the mistake:appreciating what it canactually do is both more… Last stated 2 years ago 22 Apr 2024 · 2 years ago SC Sean Carroll — holds since 2024-04-22 — tap for who they are Same subject: This incident may be the clearest warning shot we will ever get about loss of control, because the AI systems that do worse things will be much better at hiding them. — tap to centre the map on it This incident may be theclearest warning shot we willever get about loss ofcontrol, because the AI… Last stated 6 days ago 1 Sept 2026 · 6 days ago AC Ajeya Cotra — holds since 2026-09-01 — tap for who they are Same subject: The risk of humanity destroying itself WITHOUT AI is the number nobody weighs, and AI is at least as likely to be what saves us. — tap to centre the map on it The risk of humanitydestroying itself WITHOUT AIis the number nobody weighs,and AI is at least as likely… Last stated a year ago 5 Jun 2025 · a year ago LF Lex Fridman — holds since 2025-06-05 — tap for who they are Same subject: Controlling a scheming model needs no research breakthrough, which is what makes it the tractable half of the problem today. — tap to centre the map on it Controlling a scheming modelneeds no researchbreakthrough, which is whatmakes it the tractable half of… Last stated 2 years ago 7 May 2024 · 2 years ago BS Buck Shlegeris — holds since 2024-05-07 — tap for who they are Same subject: Keeping children away from AI puts them at a disadvantage, so the risk of them never learning to use it counts as much as the risk of them using it. — tap to centre the map on it Keeping children away from AIputs them at a disadvantage,so the risk of them neverlearning to use it counts as… Last stated 2 months ago 9 Jul 2026 · 2 months ago AM Adam Mosseri — holds since 2026-07-09 — tap for who they are
same subject or similar wordingbar: how recently it was last stated — full and dark this week, a faint sliver at five years

At the centre We do not control neural networks well enough to guarantee an AI will not harm humans, so we should be careful about what capabilities we give them. Last stated 17 Feb 2023 · 4 years ago Holds Armin Ronacher Read this korrent →