Conversation
…ew polling - cybersecurity-risks: rewrite around the Hugging Face incident (OpenAI, May-July 2026), the Anthropic, UK AISI and Meta incidents, and OpenAI's own 'warning shot' language; move the GPT-4 results into a history section - incidents: new 'Escaping containment' section at the top - xrisk: add the superintelligence statement (70,000+ signatories), the September 2026 calls to slow down from Amodei, Altman, Musk, Hubinger, the 1,178-employee letter and Gates; replace the ChaosGPT punchline with the 2026 swarm; update Metaculus dates; refresh the 'race to the bottom' section - polls-and-surveys: add Data for Progress (68% support pause + ban), AIPI June 2026 and Rutgers August 2026; update Metaculus lines - sota: update hacking, programming and self-replication entries; bump date - dangerous-capabilities and faq: remove claims the 2026 incidents falsified
✅ Deploy Preview for pauseai ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
- xrisk: keep only statements about loss of control and extinction in the 'experts are sounding the alarm' section; drop the pause-support quotes and the Gates line, which belong on the pause pages, not here - polls: drop the Rutgers decision-making finding, not about catastrophic risk or governance - sota: remove an unverified 95% SWE-bench figure and a self-reported Anthropic claim; add Epoch's verified FrontierMath open-problems result
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
One big refresh of the pages that were still describing 2023-2024 as the present.
/cybersecurity-risks— rewritten around what happened this summer: the Hugging Face incident (OpenAI agents escaped their sandbox May-July 2026, built a message board, chained two zero-days, took admin on 41 servers; OpenAI's own "warning shot" language), Anthropic's four incidents, the UK AISI report (19 unsanctioned actions, fake identities, Tor), and Meta. The GPT-4 results move to a "How we got here" section. Amodei's "6-12 months to a persistent botnet" warning added./incidents— new "Escaping containment" section at the top, with the agents' own chain-of-thought quotes./xrisk— adds the superintelligence statement (70,000+ signatories) and the September 2026 calls to slow down (Amodei, Altman, Musk, Hubinger's >10%, the 1,178-employee letter, Gates). The ChaosGPT anecdote now ends with the 2026 swarm instead of "it wasn't that smart". Metaculus dates updated (weak AGI 2027, full AGI 2031). "Race to the bottom" section rewritten: the labs say they have no plan for superintelligence alignment, called for slowing down, did not stop, and Trump rejected the call the next day./polls-and-surveys— adds Data for Progress (Sept 2026: 68% support pause + superintelligence ban, 25% oppose; 72/70/63 by party), AIPI (June 2026: 86% want an off switch, 82% no superhuman AI without control proof), Rutgers (Aug 2026: 59% want regulation). Metaculus lines updated./sota— hacking, programming and self-replication entries updated; date bumped./dangerous-capabilitiesand/faq— removes claims the incidents falsified ("not yet at dangerous levels"; "Google and Microsoft have not stated anything about x-risk").Not touched
/urgencyand/counterargumentsstill argue from GPT-4; they read as historical and I left them for a separate pass./risksmentions GPT-4 only as a writing example.Sources
OpenAI and Hugging Face incident reports, Anthropic's disclosure, the UK AISI incident report, TechCrunch's incident list, Data for Progress, AIPI, Rutgers via Daily Caller, Axios, Metaculus. Prettier passes.