Trump Lifts Export Controls on Anthropic Claude

Anthropic, the AI company behind the Claude family of models, builds large language systems that power chatbots, coding assistants and enterprise tools. The news driving attention this week is the U.S. government's decision to lift export controls on two of its flagship models, Claude Fable 5 and Mythos 5, after an 18 day shutdown that cut off every user worldwide.

Starting Wednesday, Fable 5 returns to Claude.ai, Claude Code, Claude Cowork and the broader Claude platform. Through July 7, subscribers on Pro, Max, Team and select Enterprise plans get access to Fable 5 covering up to half their weekly usage allotment before credits kick in. Anthropic also plans to restore the model on Amazon Web Services, Google Cloud and Microsoft Foundry, though it has not given a firm date for those rollouts.

What Triggered the Shutdown and How It Got Resolved

June 12 is when the trouble started. That's when a directive landed based on research from Amazon showing a way to coax Fable 5 into producing dangerous outputs tied to software vulnerability discovery. Because the prohibition applied to all foreign nationals, a group that even included Anthropic's own researchers, the company decided the only way to comply was to pull both models offline entirely rather than try to selectively restrict access.

Anthropic pushed back on the idea that Fable 5 was uniquely risky, saying its internal testing found that other widely used models, including OpenAI's GPT 5.5, could generate the same flagged outputs. In response, Anthropic built a new safety classifier that it says blocks the Amazon documented workaround in more than 99 percent of attempts. The Commerce Department's Center for AI Standards and Innovation reviewed both the original safeguards and the new classifier and confirmed they worked, according to Anthropic. Any Fable 5 requests that still get blocked will now be rerouted to Claude Opus 4.8.

Mythos 5 access is coming back more slowly. A limited set of U.S. organizations regained entry after the government's approval on June 26, but expanding that to more domestic and international partners through the Glasswing program is still being worked out with federal officials.

Commerce Secretary Howard Lutnick said on X that his department spent two weeks vetting Fable 5 with Anthropic to get the U.S. government aligned and to keep the country ahead in the technology race. White House chief of staff Susie Wiles echoed that sentiment, telling X, per The Wall Street Journal, that the shared goal is deploying the best technology as fast and safely as possible.

Anthropic's Valuation, Timing and Leadership Shuffle During the Standoff

The timing could hardly have been worse for Anthropic. The company recently filed a confidential IPO prospectus showing a revenue run rate of $47 billion and a valuation of $965 billion, so an 18 day blackout on two major models landed right as investors and regulators were paying close attention.

Notably, CEO Dario Amodei was not the one leading talks with Washington. CNBC reported that co founder Tom Brown took over as the primary negotiator, a shift that shows up in the record too: Lutnick's correspondence was addressed directly to Brown rather than Amodei.

Whether the New Safeguards Hold Up

With Fable 5 back online and Mythos 5 access trickling back to select organizations, the immediate crisis has passed. But the bigger question hanging over Anthropic is whether its new classifier keeps performing as well outside a controlled testing environment as it did during the Commerce Department's review, especially with an IPO process still unfolding in the background.