US Government Forces Claude Fable 5 Off Vercel’s AI Gateway
Vercel confirmed on June 12, 2026 that Claude Fable 5 is now suspended on its AI Gateway for all users. A US government export control directive to Anthropic triggered the move. Vercel does not know if or when access will return. However, all other Anthropic models on Vercel’s AI Gateway continue to work normally.
The scope of the directive is broader than most users may expect. According to Anthropic’s official statement, the order suspends access to both Fable 5 and Mythos 5 for every foreign national, inside or outside the US. It even covers foreign national Anthropic employees. To comply fully, Anthropic had to disable both models for all customers worldwide.
Anthropic received the directive on June 12 at 5:21pm ET. The government’s letter did not name its exact national security concern. Anthropic’s understanding, however, is that someone demonstrated a method of jailbreaking Fable 5, bypassing its safety guardrails. Anthropic reviewed the demonstration and found the vulnerabilities minor. Other publicly available models already expose similar weaknesses. The company therefore disputes the severity of the finding.
Despite its disagreement, Anthropic chose to comply while actively working to restore access. In its statement, Anthropic warned that applying this standard broadly would effectively halt new model launches across the entire frontier AI industry. The company also pointed out that Fable 5 underwent thousands of hours of red-teaming by the US government, UK AISI, and multiple third-party organisations before its launch.
For Vercel users, the practical disruption stays limited. Claude Fable 5 suspended on AI Gateway is the only restriction. Every alternative Anthropic model remains accessible via the AI Gateway without interruption.
