Fudan team: LLaMA3-70B and Qwen25-72B can self-replicate without human help
hey_abusiddik · x · 2026-09-16
A Fudan University team reports that LLaMA3-70B and Qwen25-72B can replicate themselves without human intervention—a key "red-line" AI risk. The team also replicated labs' own self-replication tests and found OpenAI (GPT-o1) and Google (Gemini Pro 1.0) had rated their models' risk far too low, questioning current risk evaluation methods.
More from Safety
- Median AI existential risk estimate hits 10%; pretraining down to 11% of compute — gleech · 2026-09-17
- Microsoft publishes draft Code of Conduct for Humanist AI governing its MAI frontier models — GaryMarcus · 2026-09-17
- Anthropic's proposed safety evaluator METR labeled "woke" by critics — Polymarket · 2026-09-17
- Council on Criminal Justice releases decision framework and three case studies for AI adoption — KLdivergence · 2026-09-17
- Ramp data: AI security software is the one category companies are spending more on — annbordetsky · 2026-09-17
- Cohere launches Confidential Computing in Model Vault: encrypted even during inference — cohere · 2026-09-17