UK Banks Still Denied Access to Anthropic's Mythos AI Model, Bailey Warns

Bank of England Governor Andrew Bailey said Friday that British banks remain unable to test their systems against cyber threats using Anthropic's Mythos artificial intelligence model, six weeks after concerns first surfaced. Bailey, speaking from a central banking conference in Reykjavik, attributed the delay to a political hold-up within the U.S. administration. Bailey told Bloomberg TV that Anthropic was willing to share the models on a trial basis, but the process has stalled. "It hasn't happened yet and I think this has been somewhat caught up in the process with the U.S. administration," he said, adding he could not explain why the process differed from other companies. Last month, Bailey had warned the model could "crack the whole cyber risk world open," though some cybersecurity experts have since called those fears overstated. President Donald Trump last week postponed signing a broad executive order on AI that was expected to create a voluntary framework for advanced model releases. Bailey, who also chairs the Financial Stability Board, stressed the need for a global approach to hacking risks, given the deep interconnections among banks.