After a dramatic two-week suspension triggered by US government export controls, Anthropic has officially restored access to Claude Fable 5 and Claude Mythos 5 as of July 1, 2026.
The Shutdown
Fable 5 launched on June 9, 2026, positioned as Anthropic's most capable model safe for general consumer use. Built on the same underlying technology as Mythos 5, Fable 5 came with conservative safeguards that would redirect queries on sensitive topics to the slightly less capable Claude Opus 4.8.
Three days after launch, the Trump administration issued an export control directive citing national security concerns. The trigger was a jailbreak technique flagged by Amazon researchers that could potentially bypass Fable 5's safety guardrails. The directive prohibited access to both Fable 5 and Mythos 5 for all foreign nationals, including non-US members of enterprise client companies and many of Anthropic's own employees.
The Negotiation
For two weeks, Anthropic worked with the government to address the security concerns. The company trained an improved safety classifier specifically targeting the jailbreak technique identified in the Amazon report, achieving a block rate of over 99% of cases.
The Trump administration also greenlit the return of Mythos 5 to a preapproved list of organizations, with similar staggered rollout rules as OpenAI's GPT-5.6 debut.
The Return
On June 30, the Department of Commerce lifted export controls on both models. Anthropic announced on X that access would begin restoring the following day, with global availability on Claude platforms and re-enablement on AWS, Google Cloud, and Microsoft Foundry.
"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."
— Anthropic, via X
What Fable 5 Offers
Fable 5 remains state-of-the-art on nearly all tested AI benchmarks:
- Software engineering: Stripe reported Fable 5 compressed months of engineering into days, performing a codebase-wide migration in a day that would have taken a team over two months.
- Knowledge work: Highest scores on Hebbia's Finance Benchmark for senior-level reasoning.
- Vision: New state-of-the-art for tasks involving visual analysis, including rebuilding web apps from screenshots.
- Autonomous work: Can work autonomously for longer than any previous Claude model, with improved memory and long-context capabilities.
- Drug design: Mythos 5 (the unsafeguarded version) accelerated protein design by 10x in internal testing.
Pricing remains at $10 per million input tokens and $50 per million output tokens — less than half the price of Claude Mythos Preview.
Looking Ahead
Anthropic announced new government coordination plans, including pre-release access for national security partners, rapid information sharing on jailbreaks, and collaboration with other AI labs on voluntary security standards. The company also committed dedicated teams and compute resources to support government testing and research.
The shutdown came at an inopportune time for Anthropic, which is preparing for an IPO and has been in a public feud with the government over supply chain risk design for months. The resolution marks a significant moment in the ongoing tension between AI capability and government oversight — and a precedent for how frontier model deployments may work going forward.