One reason banking professionals have heard so much enthusiasm around using generative AI is its potential financial impact on the industry.
In “Capturing the full value of generative AI in banking,” McKinsey estimates that GenAI could add the equivalent of between $200 billion and $340 billion in value annually across the banking industry. That’s equivalent to 9% to 15% of operating profits, the report notes. The greatest absolute gains forecast, largely from increased productivity, are tied to corporate and retail banking.
As generative AI adoption expands, banks and credit unions continue to have legitimate concerns about how the technology fits within a regulated environment. Financial institutions must consider whether a use case can improve a process, along with the data, controls, oversight, and technology needed to use it responsibly.
Financial institutions have several reasons to approach generative AI carefully. Three areas deserve particular attention:
1. Customer trust, privacy, and information security
Generative AI creates additional considerations around how sensitive information is handled. Financial institutions should understand what data can enter a GenAI system, where that information is stored, who can access it, whether it can be used to train underlying models, and how customer and institutional data is protected. Clear policies and access controls can help reduce inappropriate use and support customer trust.
2. Regulatory compliance, accuracy, and human oversight
Banks and credit unions remain responsible for decisions and communications supported by AI. Because GenAI can produce incomplete or inaccurate information, institutions need processes for reviewing outputs, documenting appropriate uses, and escalating issues when necessary. Explainability may also be important when AI supports decisions subject to regulatory requirements. Clear human oversight helps ensure employees understand when an output requires additional review and who remains accountable for the final decision.
3. Technology, staffing, and governance resources
Effective use of GenAI depends on the institution’s data infrastructure, technology, vendor relationships, internal ownership, policies, employee training, and capacity for ongoing oversight. Financial institution leaders should also consider the AI capabilities available through existing technology providers as they determine where additional solutions and resources may be appropriate.
Despite these considerations, financial institutions are already using generative AI in targeted ways to improve customer experiences, support employees, and increase efficiency. Examining those use cases can help banks and credit unions identify where the technology may provide value within their own operations.
