Emergency Export Restrictions Lifted on Anthropic AI Models
Late Tuesday, the Trump administration announced the removal of emergency export restrictions on Anthropic Inc.’s advanced artificial intelligence systems, bringing an abrupt end to a contentious situation that had been affecting the global tech sector, as stated by Commerce Secretary Howard Lutnick.
In a letter reviewed by the Post, Lutnick mentioned that Anthropic has committed to proactively identify and mitigate security threats linked to their models. They will work closely with the U.S. government regarding protocols and standards for their products, including Mythos and Fable, as well as future developments. Additionally, they will inform the government about any unauthorized activities.
Lutnick confirmed, “The restrictions… are withdrawn. A license is no longer required for the export, reexport, or internal transfer of Mythos or Fable models.”
This decision reverses a directive issued by the White House on June 12, which had barred foreign nationals from using Anthropic’s prominent Fable 5 and Mythos 5 systems.
Anthropic announced on X, “We have received notification that the Department of Commerce has lifted the export restrictions for Claude Fabre 5 and Mythos 5. We will begin restoring access tomorrow and will share updates soon.”
The government had previously disabled the software worldwide after the San Francisco-based startup struggled to effectively restrict foreign access in real time.
The rapid policy shift followed significant lobbying by corporate entities and behind-the-scenes discussions between Anthropic leaders and Lutnick.
Earlier reports indicated that Trump’s change of heart came after a conversation with Anthropic’s CEO, Dario Amodei, at the G7 summit in France.
This rollback is expected to provide immediate relief to the broader tech sector, which had faced operational challenges during the nearly three-week blackout.
White House Chief of Staff Susan Wiles commented, “Under President Trump’s leadership, the United States is the undisputed winner in the AI race.” She expressed gratitude to the various industries collaborating with the White House on the President’s executive order focused on advancing AI innovation and security.
Major companies heavily depend on Anthropic’s models for intricate data analysis, while software developers utilize the platform to accelerate coding processes.
The initial crackdown began following classified cybersecurity evaluations. U.S. intelligence identified that the Mythos model might reveal vulnerabilities in secure government systems, prompting officials to revise long-standing hardware-focused export laws and impose a software ban.
Technology advocates criticized the restrictions as arbitrary, noting that competing models from companies like OpenAI serve similar purposes without facing government scrutiny.
Previously, Anthropic had disputes with federal officials concerning defense contracts, as the startup declined to permit the military’s use of its tools for autonomous weapon systems. This led the Army Department to classify Anthropic as a potential supply chain threat; however, a federal judge blocked this designation in March.
Internal leaked chats highlighted frustration among Anthropic employees, with some suggesting that the company was being penalized for its strict ethical business practices.
Analysts see this costly disruption as a significant alert for Silicon Valley.
The Post reached out to Anthropic, the White House, and the Department of Commerce for further comments.
