Fable 5 Rises: Washington Unbans Anthropic's Top Models After Three-Week Standoff

The US Commerce Department lifted export controls on Anthropic's Claude Fable 5 and Mythos 5 on June 30, clearing the way for Anthropic to restore full global access to both models starting July 1. The reversal ends nearly three weeks of negotiations between Anthropic and the White House that began when the Commerce Department imposed the restrictions on June 12, three days after the models launched on June 9. Anthropic confirmed the decision in a post on X: "We've received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5. We'll begin restoring access tomorrow, and will share an update soon."

The original directive cited national security concerns reportedly linked to potential jailbreaks of the models. The rule forced Anthropic to suspend access for foreign nationals across Claude.ai, the API, AWS Bedrock and other partner platforms. The company took both models entirely offline after determining that segmenting users by nationality in real time was not operationally feasible. US Secretary of Commerce Howard Lutnick wrote on X: "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." White House Chief of Staff Susie Wiles said the government's priority remains to "get the best [AI] tech deployed as quickly and safely as possible."

Anthropic outlined its redeployment plan in a separate statement: "After a series of productive conversations with the US government, we're redeploying the model with a new set of classifiers to target and block more cybersecurity tasks." The restrictions were triggered after a report in which Amazon researchers demonstrated a method of bypassing Fable 5's safeguards, prompting the model to identify several software vulnerabilities. In a blog post, Anthropic argued this risk was not unique to Fable 5, noting that weaker models could also identify the same vulnerabilities and produce the same exploits.

Claude Fable 5 is Anthropic's most capable widely available model, built on the Mythos-class architecture and enhanced with safeguards for general use, including reasoning tasks, long-horizon agentic work, software engineering and advanced vision capabilities. Mythos 5 shares the same underlying architecture with additional safeguards for cybersecurity applications and was previously reserved for trusted partners through Project Glasswing, a collaboration announced in April with Amazon, Microsoft, Google and other partners to safeguard against AI cybersecurity threats.

Anthropic said it is drafting a consensus framework with those partners for "assessing the severity of AI jailbreaks" and is scaling up collaboration with the US government on AI model testing and safeguards, including pre-release access to models, information sharing on jailbreaks and misuse, and dedicated resources for joint research. A well-known AI researcher claimed to have jailbroken Fable 5 within 48 hours of its launch in June, before the government restrictions, and shared screenshots showing how he bypassed the model's safety guardrails.