US Treasury secretary Scott Bessent has reportedly summoned major American bank chiefs to Washington for a meeting focused on the cyber risks associated with Anthropic’s latest artificial intelligence model.
The gathering took place this week at Treasury headquarters, according to reports, following the release of Claude Mythos, which Anthropic says presents unprecedented cybersecurity risks. The meeting brought together senior figures from the banking sector as officials examined the potential implications of the new model for financial institutions and the wider system.
Among those said to have attended was Jerome Powell, chair of the Federal Reserve. His reported presence underlines the level of concern surrounding the issue, with policymakers and regulators watching closely as advanced AI tools become more capable and more widely deployed.
The reported session comes at a time when banks and regulators are increasingly focused on the security challenges posed by rapid developments in artificial intelligence. Systems such as Claude Mythos may offer powerful capabilities, but they also raise questions about how such tools could be used to support malicious activity, including cyberattacks.
Anthropic has described the model as carrying unprecedented cybersecurity risks, a statement that appears to have prompted immediate attention from government and financial leaders. While the reports do not detail specific threats discussed at the meeting, the fact that top banking executives were called in suggests the issue is being treated as urgent.
Financial institutions are natural targets for cybercriminals, and the rise of more advanced AI models has added a new layer of concern for security teams. Even without any confirmed incident linked to Claude Mythos, the meeting reflects growing anxiety about how quickly sophisticated AI capabilities can outpace existing safeguards.
The Treasury gathering also highlights the broader role of the US government in coordinating responses to emerging technology risks. With Bessent bringing bank leaders together in Washington, the administration appears to be seeking direct engagement with the industry at a moment when the intersection of AI and cyber risk is moving higher up the policy agenda.
Further details of the meeting were not provided in the source report, and it is not clear what next steps, if any, were agreed. Still, the reported attendance of Powell and other senior banking figures signals that concerns about AI-enabled cyber threats are now reaching the top levels of both government and finance.
