Global financial regulators are falling behind banks in adopting artificial intelligence, leaving them ill-equipped to monitor the risks of advanced AI models such as Anthropic’s Mythos. A new report from the Cambridge Centre for Alternative Finance indicates financial institutions are adopting AI at twice the rate of their supervisors. Only two in ten regulators report advanced AI adoption, and just 24% collect data on industry AI use. "Authorities cannot successfully harness or oversee AI if they are navigating its adoption and risks without hard data," the report states.
The research surveyed over 350 financial institutions and fintechs, 140 AI vendors, and 130 central banks and financial authorities worldwide. Regulators have increasingly warned about AI risks in finance, with Anthropic’s recent release of Mythos highlighting challenges for banking legacy systems.
Mythos exemplifies next-generation AI capable of exploiting software vulnerabilities at scale, potentially overwhelming existing human governance. While regulators maintain firms are accountable for AI-related harms, this becomes complex with autonomous, third-party AI systems. The report suggests regulators must adopt agentic AI capabilities themselves to effectively oversee advanced systems.