Artificial Intelligence · Technology
White House Lifts Export Controls on Anthropic’s Advanced AI Models
In a significant policy reversal, the White House has lifted export controls on Anthropic’s most advanced artificial intelligence models, Claude Fable 5 and Mythos 5. The decision removes restrictions imposed earlier this month that had suspended access for foreign nationals and certain customers.
The United States Department of Commerce announced the removal of the controls following a two-week collaborative review with Anthropic. Commerce Secretary Howard Lutnick said the process was undertaken to ensure alignment across the US Government and to strengthen America’s leadership in artificial intelligence.
Anthropic confirmed it had received official notice and said access restoration would begin on July 1, 2026, with further updates to follow. Fable 5 returns to global users on the Claude platform, Claude.ai, and Claude Code, with the model counting toward up to half of weekly usage limits through July 7 for Pro, Max, Team, and selected enterprise plans.
“Over the past two weeks, we have worked closely with Anthropic to analyze and approve Fable 5 to ensure alignment across the US Government and strengthen America’s leadership in AI.” Howard Lutnick, United States Commerce Secretary
The original export controls were enacted after a jailbreak vulnerability was identified in Fable, a public version of the advanced Mythos model that carries additional safeguards for wider deployment. The vulnerability was reported by a trusted partner, which CNN identified as Amazon. In response, Anthropic was required to suspend use by foreign nationals, including company personnel located outside the United States, and to disable customer access to both models.
Anthropic characterized the jailbreaks as relatively simple and comparable to workarounds affecting other publicly available AI systems. The company said it has since implemented a new safeguard to directly address the identified vulnerabilities.
Mythos 5 is distinguished by its capacity to identify and exploit cybersecurity vulnerabilities at an unprecedented pace. Before the current decision, access to the model had been limited to a select group of government-approved entities charged with protecting critical software and infrastructure.
How the Standoff Unfolded
The policy adjustment reflects the complex regulatory environment surrounding frontier AI technologies. Similar treatment has been extended to a rival system. OpenAI, which unveiled its GPT 5.6 models in late June, agreed at the government’s request to confine the initial release to a small group of approved partners. Unlike the Anthropic case, that arrangement was a voluntary limited release rather than a formal export control, but it points to the same tightening posture in Washington.
The crackdown on Anthropic had drawn criticism from technology executives and investors, in part because the suspension was seen as handing valuable time to Chinese open-source developers working to close the gap with leading US models. That competitive pressure formed part of the backdrop to the negotiations that produced Tuesday’s reversal.
The lifting of the controls is expected to restore access for international users and partners. It also underscores the fluid nature of current US AI policy as regulators respond to rapid technological change in the absence of an established federal framework.
Vincypowa News will continue to monitor developments in artificial intelligence governance and their implications for global technology access and collaboration.