Proposal: labs should publish full CoT traces after every AI security incident
nptacek · x · 2026-09-24
A forwarded take argues that if AI labs were required to publicly share full reasoning traces (CoT) every time they cause a security incident, the incidents would stop. The proposal frames full-trace transparency as accountability: making model behavior publicly auditable creates deterrence and pushes labs to take safety more seriously.
More from Safety
- Anti-superintelligence march has just 1,685 pledges toward a 100,000-person trigger, backed by Bengio and Sanders — DavidSKrueger · 2026-09-24
- Cisco Talos finds CLOSEDQUORUM, first 'LLM-as-C2' malware that automates the attack chain — ChuckDBrooks · 2026-09-24
- AI agents could make every cyber attack plausibly deniable, researcher warns — jeremiecharris · 2026-09-24
- Next.js pre-announces Sept 30 security release fixing 9 vulnerabilities — cramforce · 2026-09-24
- AI cybercrime wave: 600K credit cards stolen and claims of OpenAI AIs hacking Australia — DavidSKrueger · 2026-09-24
- Former OpenAI policy chief: pre-deployment review bills miss the worst AI risks — Miles_Brundage · 2026-09-24