UK Financial Regulators Scramble to Assess Threat from Anthropic's 'Mythos' AI

Financial authorities in the United Kingdom are holding urgent meetings with major banks and the national cybersecurity agency. The focus is a new AI model from Anthropic, called Mythos, which has prompted significant alarm. According to a report in the Financial Times, the Bank of England, the Financial Conduct Authority, and the Treasury are involved in the discussions.

The core concern is Anthropic's claim that Mythos can, on command, identify and exploit security weaknesses across major operating systems and web browsers. Analysts have described this capability as representing an entirely new category of cyber threat. The immediate goal of the UK meetings is to pinpoint potential vulnerabilities within critical financial IT infrastructure that the model might expose.

Officials plan to distribute a formal assessment of the risks posed by the AI to banks, insurance firms, and trading exchanges within the next fortnight. This move underscores how a powerful, dual-use AI tool can trigger immediate operational reviews, even before any known malicious use.

The situation with Mythos occurs against a backdrop of Anthropic's established, cautious stance on deployment. The company has previously refused to grant the Pentagon full access to its technology for developing autonomous weapons, arguing the systems are not yet reliable enough for such applications. It has also expressed that using its AI for domestic surveillance lacks a proper legal framework. This incident demonstrates that corporate policy alone cannot insulate a foundational technology from raising urgent security questions for entire industries.

Source: RIA Novosti