AI company Anthropic has suspended access to its most advanced AI models, Fable 5 and Mythos 5, after receiving a U.S. government order requiring the company to block foreign nationals from using the technology. Anthropic said the decision was made to comply with the directive, although it disagrees with the government’s reasoning and is working to restore access as quickly as possible.
According to Anthropic, officials raised concerns about a possible “jailbreak” that could bypass safeguards designed to prevent the models from being used to identify software vulnerabilities. The company said it was given only limited information and believes the issue involves a narrow risk rather than a widespread security threat. Anthropic argued that removing access to models already being used by millions of people is an excessive response.
The move marks a significant shift in U.S. AI policy. Previous restrictions mainly focused on limiting access to advanced chips and computing hardware. This order directly targets access to AI models themselves, reflecting growing concerns that powerful systems could be used to support cyberattacks or other national security threats. Experts have warned that models such as Mythos could help uncover software weaknesses if used maliciously.
The dispute comes at a sensitive time for Anthropic, which recently filed confidentially for an IPO. The company has also clashed with U.S. officials over military uses of AI, highlighting the growing tension between AI developers and regulators over how advanced systems should be governed.
Related Readings:
