Anthropic CEO Dario Amodei Rejects Call for Open‑Weight Model Ban, Warns of Chinese AI Threat

Key Points

• Anthropic CEO Dario Amodei says the company never supported a ban on open‑weight AI models.
• He differentiates safe open models from the geopolitical threat posed by authoritarian regimes, especially China.
• Amodei warns that open‑weight models could facilitate biological attacks and are hard to guardrail.
• Proposes tighter U.S. chip export restrictions and a crackdown on model distillation.
• Supports creation of a global AI safety‑testing body that includes all major AI producers, even China.