OCC Issues First AI Guidance for Bank Examiners

The Office of the Comptroller of the Currency has released its first guidance directing examiners on how to assess national banks' use of artificial intelligence.

Updated on Nov. 4, 2026 in Artificial Intelligence

OCC Issues First AI Guidance for Bank Examiners

Live Poll

Do you trust banks to use AI safely with your money?

The Office of the Comptroller of the Currency issued its first-ever guidance for examiners reviewing banks' use of artificial intelligence on Wednesday, November 4, 2026. The guidance directs examiners to focus on the consequences of potential AI-driven mistakes rather than the sophistication of the models themselves.

Why it matters

US regulators have been asking banks about their use of advanced AI models and the limits placed on them, reflecting broader concern that AI could create cybersecurity and financial stability risks. Because the guidance is not legally binding like a rule, its practical effect will depend on how examiners apply it.

The guidance covers large national banks overseen by the OCC and is explicitly non-binding, meaning it does not carry the force of a regulation. Examiners are told to weigh how a lender uses AI and the size and complexity of the institution, but specific thresholds were not stated.

The players

Office of the Comptroller of the Currency (OCC)

The OCC is the federal regulator that charters, supervises and examines national banks in the United States, and it issued this guidance to its examiners.

The details

The OCC says a simple AI agent authorized to initiate transactions could be considered riskier than a sophisticated internal model, and that higher-risk uses may require greater restrictions on autonomy or more direct human oversight. Banks are not expected to establish separate risk parameters specifically for AI use.

November 4, 2026: OCC issues first-ever AI guidance for bank examiners

Big Picture

The guidance extends the OCC's longstanding bank examination framework, which evaluates institutions on risk and oversight, to cover artificial intelligence. It follows a pattern in which regulators worldwide have begun querying lenders about how they deploy AI.

For customers of large national banks, the guidance signals that examiners will look closely at how AI systems make or execute decisions, such as transactions. Because it is non-binding, its effect on day-to-day banking services is likely to emerge gradually through supervisory expectations rather than immediately.

The takeaway

Regulators are shifting attention from how sophisticated an AI model is to what happens when it makes a mistake. For anyone tracking AI in finance, the key question is who oversees the consequences and how much human control remains in place.