conv.

← The whole story
Security
Open-weight AI models trigger year-long security crisis

Researcher reports immediate vulnerability flood from LLM audits

Datasette maintainer Simon Willison, citing a recent security audit of his project using GPT-6 Astra and Claude Fable 5.1, reports spending a full week patching extremely obscure vulnerabilities that LLMs discovered—vulnerabilities that had gone unspotted by humans for extended periods. He contradicts the one-year timeline, stating "We don't even have a year."

“I ran a security audit using GPT-6 Astra and Claude Fable 5.1 against my main open source project (Datasette) recently and then spent a full *week* fixing vulnerabilities that they found, many of which were extremely obscure, hence why nobody had spotted them before.”

simonw

apropos (author of Datasette) Security researcher, post authorSimon Willison (simonw) Datasette maintainer, open-source security expertZ.ai Co. (formerly Zhipu AI) AI model developerjfred (commenter) Security architecture discussantDeAlignAI Model abliteration group

The whole story postscomments the bright band is this development · numbered dots are the others · click one to jump

Peak 8 pieces in one hour at Sep 19, 7 PM; 58 pieces over 5 days (1 post · 57 comments) Sep 19, 2 PM — 2 pieces · 1 post · 1 comment — Lobsters 2Sep 19, 3 PM — quietSep 19, 4 PM — quietSep 19, 5 PM — 4 pieces · 4 comments — Lobsters 4Sep 19, 6 PM — 1 piece · 1 comment — Lobsters 1Sep 19, 7 PM — 8 pieces · 8 comments — Lobsters 8Sep 19, 8 PM — 1 piece · 1 comment — Lobsters 1Sep 19, 9 PM — 1 piece · 1 comment — Lobsters 1Sep 19, 10 PM — 6 pieces · 6 comments — Lobsters 6Sep 19, 11 PM — 3 pieces · 3 comments — Lobsters 3Sep 20, 12 AM — 3 pieces · 3 comments — Lobsters 3Sep 20, 1 AM — 1 piece · 1 comment — Lobsters 1Sep 20, 2 AM — quietSep 20, 3 AM — quietSep 20, 4 AM — quietSep 20, 5 AM — 3 pieces · 3 comments — Lobsters 3Sep 20, 6 AM — 1 piece · 1 comment — Lobsters 1Sep 20, 7 AM — 1 piece · 1 comment — Lobsters 1Sep 20, 8 AM — 1 piece · 1 comment — Lobsters 1Sep 20, 9 AM — 1 piece · 1 comment — Lobsters 1Sep 20, 10 AM — quietSep 20, 11 AM — quietSep 20, 12 PM — 3 pieces · 3 comments — Lobsters 3Sep 20, 1 PM — 2 pieces · 2 comments — Lobsters 2Sep 20, 2 PM — 2 pieces · 2 comments — Lobsters 2Sep 20, 3 PM — 2 pieces · 2 comments — Lobsters 2Sep 20, 4 PM — 2 pieces · 2 comments — Lobsters 2Sep 20, 5 PM — 4 pieces · 4 comments — Lobsters 4Sep 20, 6 PM — quietSep 20, 7 PM — 1 piece · 1 comment — Lobsters 1Sep 20, 8 PM — quietSep 20, 9 PM — quietSep 20, 10 PM — quietSep 20, 11 PM — quietSep 21, 12 AM — quietSep 21, 1 AM — 1 piece · 1 comment — Lobsters 1Sep 21, 2 AM — 1 piece · 1 comment — Lobsters 1Sep 21, 3 AM — quietSep 21, 4 AM — quietSep 21, 5 AM — 2 pieces · 2 comments — Lobsters 2Sep 21, 6 AM — quietSep 21, 7 AM — quietSep 21, 8 AM — 1 piece · 1 comment — Lobsters 1Sep 21, 9 AM — quietSep 21, 10 AM — quietSep 21, 11 AM — quietSep 21, 12 PM — quietSep 21, 1 PM — quietSep 21, 2 PM — quietSep 21, 3 PM — quietSep 21, 4 PM — quietSep 21, 5 PM — quietSep 21, 6 PM — quietSep 21, 7 PM — quietSep 21, 8 PM — quietSep 21, 9 PM — quietSep 21, 10 PM — quietSep 21, 11 PM — quietSep 22, 12 AM — quietSep 22, 1 AM — quietSep 22, 2 AM — quietSep 22, 3 AM — quietSep 22, 4 AM — quietSep 22, 5 AM — quietSep 22, 6 AM — quietSep 22, 7 AM — quietSep 22, 8 AM — quietSep 22, 9 AM — quietSep 22, 10 AM — quietSep 22, 11 AM — quietSep 22, 12 PM — quietSep 22, 1 PM — quietSep 22, 2 PM — quietSep 22, 3 PM — quietSep 22, 4 PM — quietSep 22, 5 PM — quietSep 22, 6 PM — quietSep 22, 7 PM — quietSep 22, 8 PM — quietSep 22, 9 PM — quietSep 22, 10 PM — quietSep 22, 11 PM — quietYesterday, 12 AM — quietYesterday, 1 AM — quietYesterday, 2 AM — quietYesterday, 3 AM — quietYesterday, 4 AM — quietYesterday, 5 AM — quietYesterday, 6 AM — quietYesterday, 7 AM — quietYesterday, 8 AM — quietYesterday, 9 AM — quietYesterday, 10 AM — quietYesterday, 11 AM — quietYesterday, 12 PM — quietYesterday, 1 PM — quietYesterday, 2 PM — quietYesterday, 3 PM — quietYesterday, 4 PM — quietYesterday, 5 PM — quietYesterday, 6 PM — quietYesterday, 7 PM — quietYesterday, 8 PM — quietYesterday, 9 PM — quietYesterday, 10 PM — quietYesterday, 11 PM — quietToday, 12 AM — quietToday, 1 AM — quietToday, 2 AM — quietToday, 3 AM — quiet 12–34–5
Sep 20Sep 21Sep 22yesterdaynow · 4:47 AM ET

What people said 2 voices · verbatim

All 5 developments of Open-weight AI models trigger year-long security crisis →

Lobsters