The Trump administration suspended foreign access to Anthropic's Fable 5 and Mythos 5 AI models on Friday following pressure from major tech firms. Amazon CEO Andy Jassy led calls for the crackdown, citing competitive concerns around advanced artificial intelligence capabilities.
The move reflects escalating tensions over AI model distribution and national security. Anthropic, the San Francisco-based AI safety company behind Claude, now faces export restrictions on its latest models. This marks a shift in how the U.S. government handles frontier AI technology, treating cutting-edge models as strategic assets requiring government oversight.
Amazon's intervention carries weight given the company's scale and Jassy's influence in Silicon Valley. The e-commerce giant operates significant cloud infrastructure through AWS and competes directly with Anthropic on AI services. Other unnamed tech firms also supported the restrictions, suggesting broader industry alignment on limiting foreign access to advanced models.
Fable 5 and Mythos 5 represent Anthropic's latest generation of language models. The suspension prevents international users and companies from accessing these systems, effectively capping Anthropic's overseas revenue streams. For a company valued at roughly $15 billion in recent funding rounds, the restriction limits growth potential outside the U.S. market.
The timing aligns with intensifying U.S.-China competition over AI dominance. Policymakers increasingly view language models and foundation models as dual-use technologies with defense implications. Earlier trade restrictions on semiconductors and chip manufacturing equipment set precedent for this approach with AI software.
Anthropic had pursued an international expansion strategy, with growing adoption in Europe and Asia-Pacific regions. The suspension immediately undermines those efforts. Competitors like OpenAI and Google may face similar pressure, though neither has announced new restrictions yet.
The decision signals the Trump administration's willingness to use national security arguments for tech policy. Unlike traditional export controls that target hardware, this targets digital goods with no marginal cost to distribute. The precedent could reshape how frontier AI companies structure their operations and go-to-market strategies.
Anthropic has not publicly commented on whether it will challenge the suspension or seek exemptions for specific markets. The company previously positioned itself as more cautious than competitors on AI safety, a stance that may complicate appeals to the administration.
This restriction represents the first major government action against a specific AI company's models post-inauguration, setting expectations for stricter tech nationalism ahead.