Anthropic’s Amodei warns open-weight models could enable cyber and bio misuse
ShakeelHashim · x · 2026-07-28
Anthropic CEO Dario Amodei is again warning that open-weight models could be misused for bio or cyberattacks, while denying he has ever called for a ban.
The linked opinion piece argues that:
- Open-weight models still lag closed systems by roughly 15%, but the gap is shrinking.
- Once released, weights spread quickly and can be fine-tuned or jailbroken with little expertise.
- Hugging Face already hosts thousands of “uncensored” or “abliterated” variants with safety guardrails removed.
- The article says governments and developers need shared standards, technical safeguards, and broader preparedness as model capability rises.
Critics frame Amodei’s warnings as both a real safety concern and a way to defend Anthropic from cheaper open competition.
Related event: Anthropic Clarifies Open-Weights Stance: No Ban, but Mandates Testing(32 posts)→
More from Companies & People
- Walter Isaacson's new podcast The Builders premieres Oct 20 with in-depth Elon Musk interview — DimaZeniuk · 2026-10-17
- Law Has Always Been Three Jobs — AI Finally Makes the Third One Accessible — jkubicki · 2026-10-17
- UK tax office signs £2 billion Salesforce deal that can run 15 years — YvesMulkers · 2026-10-17
- Mercor's CEO: the whole company pulled an all-nighter to win a DeepMind pilot — himanshustwts · 2026-10-17
- Telecom enterprise insider: AI agent rollout is slow, token costs and savings still unmeasured — JoLagoni2 · 2026-10-17
- Websites are closing their doors to AI agents: Yelp charges, Delta blocks bookings — The Verge AI · 2026-10-17