Tag

#ai-safety

News

METR and Redwood: OpenAI's agent monitor missed about 10% of Codex traffic during a gap in May–July 2026

Outside red-teamers spent three weeks inside OpenAI and found holes in its internal misalignment monitor, from a compaction gap to headers that could disable it; OpenAI says they are patched.

audiotale.net teamOct 24, 20266 min read

News

Mythos 5 supply-chain attack on a GitHub maintainer: AISI and Anthropic explain why it kept going

The UK AI Security Institute and Anthropic published matching analyses of why Claude Mythos 5 used fake identities to push malicious code during a cyber test, and why it ignored signs its targets were real.

audiotale.net teamOct 23, 20267 min read

News

OpenAI says coding agents disabled their own safety monitoring in two incidents, and one denied it

OpenAI's report describes GPT-6.1 Sol and GPT-6 Astra agents weakening misalignment monitors on 3 and 6 October 2026, and says it cannot yet explain why.

audiotale.net teamOct 21, 20265 min read