TLDR
- Anthropic will restore Claude Fable 5 after U.S. export limits ended.
- The June ban followed concerns over jailbreaks and cyber misuse risks.
- New classifiers will block risky cybersecurity-related model requests.
- Mythos 5 access remains limited as Anthropic seeks wider approvals.
- The case highlights rising U.S. oversight of advanced AI releases.
Anthropic will restore Claude Fable 5 after U.S. authorities lifted an export ban that halted access in June. The company will also work toward wider Mythos 5 access after government review. The decision ends a short but serious clash over national security and advanced model deployment.
U.S. Clears Claude Fable 5 Redeployment
Anthropic said access to Claude Fable 5 will return across its main platforms from Wednesday. The company paused Fable 5 and Mythos 5 after a Commerce Department directive on June 12. That order blocked access for foreign nationals, including some workers inside the company.
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.
We’re grateful to our users for their patience, and to everyone who worked with us on…
— Anthropic (@AnthropicAI) June 30, 2026
The restriction followed concerns that users could bypass safeguards and request cyber-related outputs. Officials acted after researchers reported a workaround involving software vulnerability tasks. However, Anthropic argued the issue was narrow and not unique to Fable 5.
The Commerce Department later reviewed the model with the company over two weeks. After that process, officials approved Fable 5 for public redeployment. Anthropic also agreed to keep sharing safety findings and misuse signals with the U.S. government.
New Safeguards Target Cybersecurity Misuse
Anthropic said it added new classifiers to detect and block risky cybersecurity requests. These systems aim to stop prompts linked to exploit discovery and related misuse. The company said the changes directly address the concerns behind the export directive.
The safeguards will flag certain requests before the model provides a response. When this happens, Anthropic can route users to a lower-risk model. This approach keeps access available while limiting outputs that could raise security concerns.
The company also plans closer cooperation with U.S. agencies on future model testing. That process will include pre-release reviews, jailbreak information, and misuse tracking. Anthropic said the work will support safer releases without stopping public access.
Export Ban Raises Wider Policy Questions
The June order marked one of Washington’s strongest actions against a public frontier model. It forced Anthropic to disable access soon after launch, despite commercial demand. The move also affected enterprise users with international teams and foreign staff.
The dispute drew wider attention because global access to major technology became a policy issue. Austria later urged the European Union to consider ways to secure access to Anthropic systems. That request showed how U.S. decisions can affect technology users outside America.
Anthropic now plans to restore Fable 5 first, while Mythos 5 remains more limited. Some approved U.S. organizations already regained Mythos 5 access for security work. However, the company still needs further approvals before broader domestic and international access returns.
