While many banks are cautious about adopting artificial intelligence, most view doing nothing as the greater risk, as they fear becoming more dependent on vendors and losing their competitive edge, according to a new survey by the American Bankers Association.
ABA surveyed banks and interviewed community bank leaders to gauge how the institutions are navigating the shift to AI. It found that the industry is actively engaging with the technology, “but doing so unevenly, cautiously and with a strong emphasis on governance.”
The key findings:
- Bankers consistently warned that avoiding AI adoption entirely leads to greater vendor dependence, erosion of internal expertise, and loss of competitiveness as AI becomes embedded in core banking operations.
- Banks draw a sharp distinction between traditional AI — including automation, pattern recognition and analytics — and generative AI. Traditional AI is relatively mature, while gen AI remains early stage and tightly controlled, with significantly lower confidence in near-term benefits due to data security, accuracy, and regulatory concerns.
- Responsibility for AI strategy rarely sits with IT alone, with leadership and other parts of the organization engaged primarily through risk committees.
- Banks are prioritizing low-risk applications that support human judgment — such as policy analysis, compliance testing, document review, research, and internal communications — rather than customer-facing or autonomous decisioning.
- Where AI is in use, banks report meaningful productivity gains in information-heavy workflows.
- Data readiness, governance, regulatory uncertainty and information security are the most commonly cited challenges in AI adoption.
