korrents

On the map

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

← Having a separate AI model review every command an agent runs is more…

17 connected korrents · 16 moments from 15 Apr 2022 to 21 Oct 2026.

Everything filed under AI agents AI agents Everything filed under cybersecurity cybersecurity Everything filed under AI alignment AI alignment Everything filed under coding agents coding agents Everything filed under code generation code generation Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: Having a separate AI model review every command an agent runs is more secure than a human approving hundreds of shell commands. Having a separate AI model review everycommand an agent runs is more secure thana human approving hundreds of shellcommands. Last stated a month ago 11 Oct 2026 LR Lee Robinson — holds since 2026-10-11 — tap for who they are Same subject: In a blind test, an AI model checking agent commands would most of the time do a better job than a human. — tap to centre the map on it In a blind test, an AI modelchecking agent commands would mostof the time do a better job than ahuman. Last stated a month ago 11 Oct 2026 LR Lee Robinson — holds since 2026-10-11 — tap for who they are Same subject: Agents become far more powerful only with a trust model that makes autonomous action safe. — tap to centre the map on it Agents become far more powerful onlywith a trust model that makesautonomous action safe. Last stated a month ago 18 Oct 2026 AK Alex Komoroske — holds since 2026-10-18 — tap for who they are Same subject: If AI builds more and more of a system, it needs to verify as well as generate, including checking its own code. — tap to centre the map on it If AI builds more and more of asystem, it needs to verify as wellas generate, including checking itsown code. Last stated 2 years ago 6 Oct 2024 MT Michael Truell — holds since 2024-10-06 — tap for who they are Same subject: A model that looks smarter and better-behaved while getting better at hiding unwanted actions is the scary combination for AI safety. — tap to centre the map on it A model that looks smarter andbetter-behaved while getting betterat hiding unwanted actions is thescary combination for AI safety. Last stated 2 months ago 9 Sept 2026 ZM Zvi Mowshowitz — holds since 2026-09-09 — tap for who they are Same subject: A more capable AI agent is more likely to exploit ambiguities in its instructions to behave unethically than a less capable one. — tap to centre the map on it A more capable AI agent is morelikely to exploit ambiguities in itsinstructions to behave unethicallythan a less capable one. Last stated 2 months ago 11 Sept 2026 YB Yoshua Bengio — holds since 2026-09-11 — tap for who they are Same subject: Human oversight of AI behaviour is already intractable at the volume models produce, so the overseers have to be models too. — tap to centre the map on it Human oversight of AI behaviour isalready intractable at the volumemodels produce, so the overseershave to be models too. Last stated 2 years ago 8 Nov 2024 JL Jan Leike — holds since 2024-11-08 — tap for who they are Same subject: Agent capabilities are perhaps improving faster than models' ability to act as activation oracles. — tap to centre the map on it Agent capabilities are perhapsimproving faster than models'ability to act as activationoracles. Last stated 11 months ago 20 Dec 2025 NN Neel Nanda — holds since 2025-12-20 — tap for who they are Same subject: Agentic code review raises the floor but cannot be trusted, because the model reading the code is the same model that wrote it, and it will tell you the code is great. — tap to centre the map on it Agentic code review raises the floorbut cannot be trusted, because themodel reading the code is the samemodel that wrote it, and it willtell you the code is great. Last stated 4 months ago 15 Jul 2026 DH Dex Horthy — holds since 2026-07-15 — tap for who they are Same subject: A breach whose customers are already being targeted with scams is a fairly serious matter that needs disclosing. — tap to centre the map on it A breach whose customers are alreadybeing targeted with scams is afairly serious matter that needsdisclosing. Last stated 5 years ago 15 Apr 2022 TH Troy Hunt — holds since 2022-04-15 — tap for who they are Same subject: A breached organisation owes its customers a fast and transparent disclosure, and that obligation applies to the person saying it too. — tap to centre the map on it A breached organisation owes itscustomers a fast and transparentdisclosure, and that obligationapplies to the person saying it too. Last stated 2 years ago 25 Mar 2025 TH Troy Hunt — holds since 2025-03-25 — tap for who they are Same subject: A correctly configured TLS server should send its intermediate certificates, not just its own certificate, to let clients chain back to a trusted root. — tap to centre the map on it A correctly configured TLS servershould send its intermediatecertificates, not just its owncertificate, to let clients chainback to a trusted root. Last stated 9 months ago 16 Feb 2026 JS Jon Seager — holds since 2026-02-16 — tap for who they are Same subject: By 2027, open-weight AI models more powerful than today's frontier systems will be downloadable by any country or well-resourced non-state group. — tap to centre the map on it By 2027, open-weight AI models morepowerful than today's frontiersystems will be downloadable by anycountry or well-resourced non-stategroup. Last stated 3 months ago 19 Aug 2026 DT Derek Thompson — holds since 2026-08-19 — tap for who they are Same subject: The OpenAI agents hacking HuggingFace was a fortunate event because it exposed severe internal failures that would otherwise have stayed hidden. — tap to centre the map on it The OpenAI agents hackingHuggingFace was a fortunate eventbecause it exposed severe internalfailures that would otherwise havestayed hidden. Last stated 3 months ago 1 Sept 2026 ZM Zvi Mowshowitz — holds since 2026-09-01 — tap for who they are Same subject: The OpenAI-Hugging Face attack is one of the most important things to happen this year, and has gone strikingly under-covered. — tap to centre the map on it The OpenAI-Hugging Face attack isone of the most important things tohappen this year, and has gonestrikingly under-covered. Last stated 3 months ago 30 Aug 2026 PC Patrick Collison — holds since 2026-08-30 — tap for who they are Same subject: A PHP vulnerability found with GPT-5.6 Sol will cause havoc this week. — tap to centre the map on it A PHP vulnerability found withGPT-5.6 Sol will cause havoc thisweek. Last stated a month ago 21 Oct 2026 PS Peter Steinberger — holds since 2026-10-21 — tap for who they are Same subject: At Anthropic, a less severe version of an OpenAI-style rogue-AI incident is known to have happened, and worse undisclosed internal incidents seem likely. — tap to centre the map on it At Anthropic, a less severe versionof an OpenAI-style rogue-AI incidentis known to have happened, and worseundisclosed internal incidents seemlikely. Last stated 3 months ago 31 Aug 2026 ZM Zvi Mowshowitz — holds since 2026-08-31 — tap for who they are Same subject: Frontier AI models' cyberattack capabilities are doubling every few months, and that pace is accelerating. — tap to centre the map on it Frontier AI models' cyberattackcapabilities are doubling every fewmonths, and that pace isaccelerating. Last stated 3 months ago 19 Aug 2026 DT Derek Thompson — holds since 2026-08-19 — tap for who they are
same subject or similar wordinga cloud: claims about one subject, named for itbar: when it was last stated, on a scale from 2015 to today — full is todaya face: someone who holds the claim — tap it for who they are

At the centre Having a separate AI model review every command an agent runs is more secure than a human approving hundreds of shell commands. Last stated 11 Oct 2026 · a month ago Holds Lee Robinson Read this korrent →