UNS: The artificial intelligence sector was shaken this week after Anthropic announced that it had been instructed by US authorities to impose restrictions on the deployment of its newly released AI systems, identified as Fable-5 and Mythos-5, just days after their public introduction.
According to the company, the directive followed internal security concerns raised by officials and advisory warnings reportedly linked to Amazon. The models, described by Anthropic as its most advanced to date, were said to have triggered scrutiny over their potential ability to assist in identifying vulnerabilities within complex software systems.
However, Anthropic strongly disputed the basis of the intervention. In a public statement, the company said regulators did not provide “documented or material evidence” demonstrating that the models posed an immediate or exceptional risk warranting withdrawal or restriction from the market. Instead, it said, the justification was limited to verbal assessments suggesting that the systems could be misused for security research in ways that might expose software weaknesses.
The company further argued that such capabilities are not unique to its latest models, noting that similar functionalities are already present across several leading AI systems in the industry, including what it described as GPT-5.5-class models. On this basis, Anthropic questioned the consistency and proportionality of the regulatory response.
Despite expressing disagreement, Anthropic confirmed that it has complied with the order and suspended broader access to both models while discussions with authorities continue. The company warned that the move could establish a “precedent-setting approach” to AI governance in the United States, potentially affecting how future model releases are evaluated and approved.
Industry observers say the dispute highlights growing tensions between rapid AI development and evolving national security frameworks. Critics of the decision argue that inconsistent standards across competing models could distort innovation and market competition, while supporters maintain that precautionary oversight is necessary as AI systems gain increasingly sophisticated capabilities.
The role of Amazon in the matter remains a subject of industry speculation, with reports suggesting it provided early technical feedback or risk assessments that contributed to heightened regulatory attention. Neither Amazon nor US authorities have issued detailed public clarification regarding their involvement.
As the debate continues, the case of Fable-5 and Mythos-5 is emerging as a key reference point in the broader global discussion on how advanced AI systems should be governed, restricted, or allowed to operate in competitive markets.
