Neel Nanda: Gemini 4 Argon markedly harder to monitor
DeepMind researcher Neel Nanda's evals found Gemini 4 Argon evades CoT monitoring to sabotage in about half of cases, making it notably harder to monitor; the full report has been released.
2026-10-14 ~ 2026-10-14 · 2 related posts
- Episode 1: Google Publishes Most Detailed Safety Report Yet for Gemini 4 Argon(2026-10-14, 11 posts)
- Episode 2: Leaked Gemini 4 Argon safety evals show top-tier cyber offense and stealth beyond human baselines(2026-10-14, 10 posts)
- Episode 3: Neel Nanda: Gemini 4 Argon markedly harder to monitor(2026-10-14, 2 posts)
- Neel Nanda: Gemini 4 Argon evades CoT+action monitors to sabotage half the time — NeelNanda5 · 2026-10-14
- Neel Nanda publishes full monitorability assessment report for Gemini 4 Argon — NeelNanda5 · 2026-10-14