https://t.co/RxrYwhoobg https://t.co/lsO0ERjvWT
SemiAnalysis Twitter · SemiAnalysis (@SemiAnalysis_) · 2026-07-27
Anthropic CEO Dario Amodei clarifies that Anthropic has never advocated for a ban on open-weights models, while outlining specific national security concerns about authoritarian AI dominance and calling for chip export controls, anti-distillation enforcement, and mandatory safety testing.
Appears in
Extraction
Topics: open-weights-modelsai-policyai-safetychip-export-controlsnational-security
Claims
- Anthropic has never advocated for a ban on open-weights models and considers open-weights models without dangerous capabilities a public good.
- The primary national security concern is authoritarian governments building AI models more powerful than those from the US and using them for military superiority or deep surveillance.
- Open-weights models pose higher misuse risks than closed models because guardrails cannot be applied and weights cannot be withdrawn once released.
- Chip export controls and crackdowns on distillation operations are more targeted and effective security measures than banning open-weights model use by US businesses.
- All sufficiently capable models, open and closed, should undergo mandatory safety testing for cyber, biological, and alignment risks before release.
Key quotes
Anthropic has never advocated for a ban on open-weights models.
Open-weights models that don't have dangerous capabilities are a public good: they don't cost anything besides the compute needed to run them, and they provide value to businesses, developers, and researchers.
Banning the use of these models by US businesses does nothing to address this risk, because bad actors are unlikely to be legitimate US businesses. It would protect US AI companies from competition, but that has never been my goal.