The US banking regulator has issued its first-ever guidance on how examiners should assess banks' use of artificial intelligence (AI), signalling a more structured approach to monitoring the technology as its adoption expands across the financial sector.
Guidance from the Office of the Comptroller of the Currency (OCC), the agency responsible for national banks, was released on Wednesday, November 4th.
The framework calls for the examination of how banks utilize AI but stresses that it is not supposed to fall into the distinct category of risk. This follows the global regulatory focus on risks related to AI, especially cybersecurity and financial stability risks.
Under the new framework, banks are not expected to establish separate risk parameters specifically for artificial intelligence. The OCC has also advised examiners against treating AI as a unique risk category.
Instead, supervisors should assess how individual lenders use the technology, taking into account their size and operational complexity.
The guidance is not legally binding in the same way as a formal regulation. However, banks generally follow such regulatory guidance closely.
The OCC said examiners should focus on the potential consequences of AI-related errors rather than the technical sophistication of the models being deployed.
This approach places greater emphasis on how AI systems function within banking operations and the impact they could have if something goes wrong.
Guidance from the OCC stresses that the risks of using AI technologies are contingent on the duties that are allocated to it.
This means that, for example, a fairly basic AI technology that has the right to carry out financial transactions can be viewed as riskier than an advanced AI tool which is intended for internal use only.
This demonstrates that the degree of autonomy and the consequences of actions taken by AI technology have to matter more in regulatory considerations than the complexity of the technology itself.
Also Read: How Agentic AI is Changing Enterprise Software Development
This OCC guidance emerges at a time when there is increasing international concern about the ramifications of AI integration in financial institutions.
Global banking regulators have been probing the use of AI by banks, the constraints on such AI and the methods used to control their activities.
In the US, regulators have also been looking into the use of complex AI models by banks and the risk management policies in place.
Given the increasing integration of AI in the finance sector, the latest guidance provides examiners with a guide in assessing AI use without classifying it separately as a risk.