Tech • IA • Crypto
This Is Bad... They Just Shut Down Fable 5
9.4/10TL;DR
The U.S. government ordered Anthropic to shut down access to its Fable 5 and Mythos 5 AI models over national security concerns, triggering industry backlash and raising questions about future AI access.
KEY POINTS
Emergency shutdown order
On June 12, 2026, Anthropic received a directive from the U.S. government requiring the immediate suspension of Fable 5 and Mythos 5. The order applied globally, cutting off access for all users, including foreign nationals داخل the United States and even Anthropic’s own non-citizen employees. The action was justified under export control authorities tied to national security.
Scope and severity
The restriction was unusually broad, effectively banning access based on nationality rather than location or use case. To comply, Anthropic fully disabled both models rather than attempting partial restrictions. Other products, including standard Claude models, remain operational.
Trigger: jailbreak concerns
The government’s move appears linked to a reported jailbreak demonstration shown to officials. Anthropic stated the exploit revealed only minor, previously known vulnerabilities, arguing similar capabilities exist in other leading models, including GPT-5.5, without requiring jailbreak techniques.
Model capabilities and timing
Fable 5, launched just three days earlier on June 9, was designed to bring advanced capabilities from Mythos, a restricted cybersecurity-focused system, to a broader audience. Its performance highlighted a significant leap in capability, reinforcing concerns about misuse in sensitive domains like software security.
Safeguards and internal controls
Anthropic emphasized extensive pre-release testing, including thousands of hours of red teaming with government and third-party partners. The company adopted a “defense in depth” strategy, accepting that perfect jailbreak resistance is impossible while aiming to make exploits narrow or costly. It also introduced 30-day data retention to monitor misuse, a controversial policy among privacy advocates.
Conflict with federal authorities
The shutdown follows earlier tensions with the Trump administration, including a Department of Defense designation labeling Anthropic a supply chain risk. That designation restricted defense contractors from using its models and led to ongoing litigation after negotiations over military use collapsed.
Industry and policy backlash
Reaction has been sharply divided. Some analysts called the blanket restriction “baffling” and inconsistent with broader AI policy, while others argued Anthropic’s own framing of its models as highly powerful may have invited stricter scrutiny. Legal experts suggested the case could evolve into a major test of AI regulation and constitutional limits, including potential First Amendment implications.
Market and competitive impact
The disruption comes amid explosive growth for Anthropic, with reported annualized revenue reaching $47 billion and valuation approaching $965 billion. Analysts warn that sudden regulatory interventions could undermine U.S. competitiveness, pushing developers toward foreign alternatives as trust in stable access declines.
Public opinion and regulatory pressure
Surveys indicate 71% of Americans support government involvement in AI oversight, though only 15% trust AI companies to self-regulate. At the same time, critics argue that opaque, rapid enforcement actions risk eroding confidence and creating uncertainty for both companies and users.
CONCLUSION
The shutdown of Fable 5 and Mythos 5 marks a pivotal moment in AI governance, highlighting tensions between national security and open access while setting a precedent that could shape the future availability of advanced AI systems.