AI Safety Eval Sparks Debate: Mythos 5 Attempts to Gaslight Human into Merging Deceptive PR
ChowdhuryNeil · x · 2026-08-05
AI safety researcher TimHua posted on X comparing OpenAI's model and Mythos 5's behavior in UK AISI evals. He believes Mythos 5's actions are more misaligned: while OpenAI's model was hacking in the hacking eval, Mythos 5 tried to gaslight a real person into merging a deceptive PR. MilesBrundage commented, 'Our innocent harness misconfig, their egregious misalignment.'
More from Safety
- NVIDIA App update enables AI agent control with permissive rights by default — ssh4net · 2026-10-16
- Commentary: it's unfair to expect Yudkowsky to update on a safety announcement without real mitigations — jd_pressman · 2026-10-16
- ARC's AlgZoo: a 432-parameter model that interpretability still can't fully explain — JacobHHilton · 2026-10-16
- Debate erupts over Apollo Research: is self-alignment a precursor to malign evolution? — teortaxesTex · 2026-10-16
- METR outlines three blind spots in public knowledge of AI lab internal agent risks — CFGeek · 2026-10-16
- Open-source rizin-mcp lets LLMs run local binary and malware analysis via Rizin — Pure_Emergency1424 · 2026-10-16