Consumer-facing functions like collections and recovery and over trust of AI may expose banks to risk
Wolters Kluwer Compliance Solutions launched a new report today analyzing the maturity of US-based banks’ AI use, governance and risk mitigation. The “US Banking AI Risk and Governance Index” analyzes survey responses from 230 US banking professionals – spanning community banks under $1 billion AUM to large institutions above $50 billion – to understand the rapidly shifting AI landscape from the leaders who are accountable for the technology’s successful implementation.
"As AI adoption has quickly shifted from a future plan to a must-have, AI governance remains an elusive target as banks face increased pressure to leverage the technology,” said Aoife May, Product Strategy Associate Director at Wolters Kluwer Financial & Corporate Compliance. “We are seeing a clear gap between how fast financial institutions can roll out AI and the accountability and transparency that their consumer base demands. These organizations need to have clear oversight and guardrails in place, including humans-in-the-loop, to ensure AI can unlock business value instead of opening them up to increased scrutiny."
The report clearly shows that banks understand the importance of investing in AI with the right tools, but there are opportunities to build out organizational governance structures, consumer protection infrastructure and organizational best practices. In fact, over 36% of respondents note that model governance and validation are their biggest risk factor when scaling AI. Additional key findings from the “US Banking AI Risk and Governance Index” include:
Automation may affect customers when most vulnerable
- Collections are often one of the more volatile areas of banking, as a customer may be in financial distress and due to the lack of regulatory oversight. This rings true as more than a third of respondents (35.22%) note collections and recovery as the banking function that introduces the highest risk of consumer harm or regulatory exposure.
- This was significantly higher than the next response of credit risk and underwriting at 25.22% – as this area operates within more developed landscape.
- When it comes to risk stemming from agentic AI, lending and underwriting workflows (33.04%) as well as collections and recovery (30.43%) were the clear leading responses – further signaling the potential issues with AI use during consumers most vulnerable financial moments.
The technology is sound, but humans may trust it too much
- A staggering 72% of respondents said that their institution is least prepared for reporting or shutting down an AI incident – with 37.83% saying regulatory reporting of an AI failure is what they are least prepared for, along with another 34.35% saying model kill-switch protocols aren’t in place.
- Behavioral and cultural risk factors made up the largest segments of human-centric risk factors with AI use. With 34.21% of respondents noting automation bias and then 27.19% stating misalignment of incentives as posing the greatest risk to their AI safety framework.
Risk mitigation paces AI adoption
- The biggest driver of safe AI adoption at survey respondents’ institutions was risk mitigation at 36.52% – outpacing regulatory compliance at 30%.
To learn more about the pressures US banks are facing with the use of AI, download the full “US Banking AI Risk and Governance Index.”