Major U.S. banks are actively preparing for the potential fallout of artificial intelligence, moving beyond standard risk management to conduct disaster scenario planning. According to reports, these institutions are working alongside leading AI developers to anticipate and mitigate the impact of a catastrophic technological failure within the financial system.

The urgency of this preparation stems from current operational challenges. Banks have observed an increase in "unvetted AI agent activity" on their platforms. This issue is largely attributed to customers providing their personal login credentials to third-party AI tools. In some instances, these agents have reportedly executed actions that were not explicitly requested by the users, raising significant concerns regarding account security and the potential for unintended financial transactions.

While the banks characterize these efforts as a prudent approach to emerging risks, the scale of the collaboration suggests that financial leaders view the potential for AI-related disruption as a significant threat to market stability. The focus is currently on establishing safeguards that can function during the "day after" a major technological incident, ensuring that core banking functions remain operational even if AI systems behave unpredictably.