UK financial regulators are in urgent discussions with big banks and the government’s cybersecurity agency to evaluate the risks associated with Anthropic’s latest artificial intelligence model.
According to the newspaper, representatives from the Bank of England, the Financial Conduct Authority, and HM Treasury are in discussions with the National Cyber Security Centre to investigate possible weaknesses in vital IT systems brought to light by Anthropic’s most recent AI model.
It stated that representatives of major British banks, insurers, and exchanges are anticipated to be told of the cybersecurity dangers posed by the model, Claude Mythos Preview, at a meeting with regulators in the next two weeks.
The action comes after big Wall Street banks and U.S. Treasury Secretary Scott Bessent met to discuss the model’s potential for cyber danger.
According to the AI firm, the model is being used as part of “Project Glasswing,” a regulated programme that allows certain organizations to employ the Claude Mythos Preview model for defensive cyber protection.
The business claimed earlier this month in a blog post that the model has already found “thousands” of significant vulnerabilities in web browsers, operating systems, and other popular software.
