On June 12, the U.S. government pulled one of the most powerful AI models available to the public offline. For eighteen days, Anthropic’s Fable 5 sat beyond reach for users around the world. That changed when the Commerce Department lifted its export controls, with Fable 5 returning across Claude tiers and platforms.
The return comes with new guardrails and a commitment giving the U.S. pre-release access to Anthropic’s future models. That is a meaningful shift in how frontier deployments are managed, and it signals Washington has moved from occasional intervention to a standing seat at the table.
The updated safety filter is designed to block the cybersecurity issue that triggered the original order over 99% of the time. When it does trip, users now receive a clear notice and a fallback answer from Opus 4.8. Anthropic has warned the filter can also flag harmless coding and debugging requests, although the company says the vast majority of coding work is unaffected.
Paid plan users are currently capped at half their weekly limits until July 7, after which access returns via usage credits. Free tier access is restored as well.
Why this matters
After eighteen days offline, Anthropic’s frontier model is back, but with a different political and technical operating environment. The pre-release access agreement raises the stakes for every future Anthropic rollout, because Washington’s review now looks like a standard step rather than an emergency brake.
At the same time, OpenAI’s GPT-5.6 is expected this week. That timing makes the next few days unusually important. We may soon see whether Google or OpenAI face similar export pressures, or whether Anthropic’s new arrangement becomes the template everyone else rejects.
Frontier AI is no longer just a product race. It is a regulatory relationship, and Anthropic just set the terms for the next era.