The US OCC gives bank examiners a framework for reviewing AI use: no separate AI risk category, with scrutiny scaled to what could go wrong
On November 4 the Office of the Comptroller of the Currency (OCC), which supervises US national banks, gave its examiners a framework for reviewing how banks use AI, which the document calls "Super Intelligence." Examiners should not treat AI as a separate risk category, and banks need no separate AI risk appetite; its risks fit into existing areas such as operational, compliance and credit risk. A use case's risk tier can rest on the consequences of failure, how much authority the system is given and how much human oversight remains, not on how advanced the model is. The 16 components, from permitted use to vendor risk, cybersecurity and testing, mainly target generative and agentic AI. The OCC says it sets no new requirements for banks and will be updated as the technology changes.
Why it matters
It shows how the main supervisor of US national banks will examine bank use of AI, especially agents that can act on their own, which shapes how banks deploy and control these systems.
Sources1
Written by dalao.ai: the title, summary and why it matters are our own words. The original reports are the record; their publishers keep authorship.