Anthropic said it will begin restoring Fable access globally on Wednesday across its own platforms and will reenable the model on services including Amazon Web Services and Google Cloud “as quickly as possible.” The company also said it will continue working with the U.S. government to expand availability of its more powerful Mythos model to select partners in the United States and other markets.
“We’re also strengthening our level of collaboration with the US government on new pre-release testing, information sharing, and research collaboration,” Anthropic wrote in a blog post. “Our hope is that this collaboration, along with our proposed consensus industry framework, will serve as the basis for systematic rules for the whole industry.”
The move reverses restrictions introduced on June 12 that blocked Anthropic from providing access to both Fable and Mythos. The government later eased limits on Mythos for select partners, while negotiations over restoring Fable continued before Tuesday’s announcement.
Anthropic said it has also added new technical safeguards designed to prevent users from circumventing Fable’s cybersecurity protections. The company said it has been discussing a technical assessment framework with the government that could provide a standardized way to evaluate similar risks in future frontier AI models.
Fable is Anthropic’s public-facing version of the Mythos model and includes additional protections intended to reduce misuse. The export restrictions had effectively halted public access because complying with licensing requirements for foreign users proved impractical at scale.
The reversal follows weeks of negotiations between Anthropic and the Commerce Department and comes amid broader debate over how the United States should oversee advanced AI systems without slowing their deployment. The administration has faced pressure to balance national security concerns with maintaining the competitiveness of U.S. AI companies as international rivals release increasingly capable models.
The policy shift also comes after OpenAI agreed to a limited rollout of one of its latest models following discussions with the White House. According to the original reporting, OpenAI’s GPT-5.6 was made available only to a small group after requests from the White House, the Office of Science and Technology Policy, and the Office of the National Cyber Director, adding to industry uncertainty over how future frontier AI models will be reviewed before release.
The White House has pointed to President Donald Trump’s June executive order emphasizing voluntary federal reviews of advanced AI systems. In a post on X, White House chief of staff Susie Wiles wrote, “Our shared priority remains: get the best tech deployed as quickly and safely as possible.”
Despite the lifting of restrictions, industry groups say questions remain about how similar decisions will be handled in the future.
“It is a positive development that Commerce has lifted the export controls on Mythos and Fable, but there remains a real need for a consistent process and framework for frontier model assessment,” said Paul Lekas, head of global public policy and government affairs at the Software & Information Industry Association.
Dean Ball, a former AI adviser in the Trump White House, also questioned what standards had been applied. “We have no idea what Anthropic did to make the models ‘safe,’ what commitments Anthropic has made going forward, and whether or how any of this applies to other frontier models in the government’s licensing queue,” he wrote in a social media post.
Joe Hoefer, chief AI officer at Monument Advocacy, said the agreement should not be viewed as a permanent policy framework. “This was resolved through direct engagement and a negotiated set of commitments, which makes sense given how new this territory is. But it’s still a case-specific agreement, not a standard the next company can point to.”
Hoefer added that the outcome represents “more of a ceasefire” than a lasting solution. “Until there’s a codified process with defined triggers and a real evidentiary standard, every frontier model launch carries the risk of a repeat,” he said.
Glenn Gerstell, former general counsel of the National Security Agency, argued the administration reached the right outcome. “Rescinding the export restrictions on Anthropic’s new models is the right step,” Gerstell said. “We’re not going to maintain our lead over China in cutting-edge AI simply by slapping export controls on the latest innovation. The whole episode underscores how we have a lot more work to do to figure out the right way of balancing regulation for safety and national security reasons with promoting innovation in the AI sector.”
This analysis is based on reporting from Politico.
Image courtesy of SQ Magazine.
This article was generated with AI assistance and reviewed for accuracy and quality.
