The US government has ordered Anthropic to shut down access to two of its most powerful AI models, Claude Fable 5 and Mythos 5, citing national security concerns. The directive, which was received by Anthropic on Friday at 5:21 pm ET, requires the company to disable both models for all users worldwide, not just foreign nationals as initially intended.
Anthropic has complied with the order, but expressed its disagreement with the government's decision in a blog post. The company argues that the finding of a narrow potential jailbreak should not be cause for recalling a commercial model deployed to hundreds of millions of people. Anthropic claims that the workaround is likely also present in AI systems developed by other companies and that it has implemented strong safeguards to reduce the likelihood of misuse.
Background and Context
Fable 5 was released on Tuesday, trumpeting the new model as the most powerful AI system Anthropic had ever shared. The company said that these AI models are powerful enough to merit strict safety guardrails to prevent malicious or dangerous use. Mythos 5, on the other hand, is a version of the company's Mythos AI model with safeguards that prevent it from answering questions about cybersecurity, biology, and chemistry.
Anthropic had opted not to make Mythos Preview generally available, instead limiting its rollout to a small group of organizations as part of a cybersecurity initiative called Project Glasswing. The goal was to give companies and organizations an opportunity to use its powerful cybersecurity capabilities to improve their defenses, and stem concerns that the technology could be exploited by bad actors to develop powerful hacking tools.
Why it Matters to the Industry
The government's decision to shut down access to Fable 5 and Mythos 5 has significant implications for the AI industry. Anthropic is widely expected to pursue an IPO this year, and its handling of Mythos has been a major point of contention. The company had promoted Mythos as a model so dangerous it couldn't be released publicly, but ultimately decided to launch a controlled program called Project Glasswing.
The incident highlights the challenges faced by AI companies in balancing innovation with safety and security concerns. Anthropic's decision to disable access to Fable 5 and Mythos 5 may have been seen as a necessary step to comply with government regulations, but it also raises questions about the effectiveness of these safeguards.
What Comes Next
Anthropic is expected to continue working with the US government to resolve the situation. The company has stated that access to all other Anthropic models would remain unaffected, and that it was working urgently to resolve the issue. However, the incident may have a lasting impact on the AI industry as a whole.
The government's decision to shut down access to Fable 5 and Mythos 5 sends a clear message about the importance of safety and security in AI development. It remains to be seen how this will affect Anthropic's plans for an IPO, but one thing is certain: the incident has sparked a renewed debate about the role of government regulation in the AI industry.
Key Facts
- The US government ordered Anthropic to shut down access to Fable 5 and Mythos 5 citing national security concerns.
- The directive requires Anthropic to disable both models for all users worldwide, not just foreign nationals as initially intended.
- Anthropic has complied with the order but expressed its disagreement with the government's decision in a blog post.
- Fable 5 was released on Tuesday and is a version of Mythos with safeguards that prevent it from answering questions about cybersecurity, biology, and chemistry.
- The company had opted not to make Mythos Preview generally available, instead limiting its rollout to a small group of organizations as part of a cybersecurity initiative called Project Glasswing.
