NAFCU Journal November December 2021

8 THE NAFCU JOURNAL November–December 2021 WASHINGTON AND INDUSTRY BRIEFS AN OVERVIEW OF AI A s technology evolves and rapidly changes the world, credit unions must shift as well. Here’s an overview of everything to know on the implementation, regu- lation and leveraging of artificial intelli- gence (AI) technologies within the credit union industry. What is AI? In Stuart Russell and Peter Norvig’s Berkeley study, Artificial Intelligence: A Modern Approach, they navigate four ways to approach the concept of AI, keeping a focus on ethics in technological modernization. Human approach: ■ Systems that think like humans ■ Systems that act like humans Ideal approach: ■ Systems that think rationally ■ Systems that act rationally In simple terms, AI is a subfield within the tech industry that combines com- puter science with data and algorithms to create systems that can problem solve on their own. Seemingly, daily use of AI can range from listening to a curated playlist on the music streaming app Spotify, to a connected computer system used by all employees of an organization. What Does it Mean for Credit Unions? AI adoption can be highly beneficial for credit unions because it can help the organization to remain highly responsive to members by streamlining member service interactions, yield new insights about member finances and improve lending operations. NAFCU remains mindful of AI’s cost savings potential as integration of alternative data and expan- sion of digital banking at credit unions becomes more common. At NAFCU’s 2021 Congressional Caucus, an AI panel covered the overall implementation of artificial intelligence into credit unions, combatting fraud and managing heavy amounts of data through cost limitation and system infrastructure. For credit unions, AI helps with multiple credit union constants, including: ■ Compliance requirements; ■ Loan growth; ■ Cybersecurity; and ■ Membership growth. In March of this year, federal financial regulators issued a request for informa- tion (RFI) seeking insights into financial institutions’ use of AI and machine learn- ing. Specifically, the RFI sought input on: ■ How AI technologies are being used for consumer services and busi- ness purposes; ■ Whether agencies should clarify any aspects concerning the use of manage- ment of AI to enable financial sector participants to deploy AI in a safe and sound manner and in compliance with consumer protection regulation; and ■ How financial institutions are addressing changes associated with AI “explainability”—or the conceptual soundness of the underlying model— and what methods exist for validating the soundness of AI output. Where Does NAFCU Stand? NAFCU recognizes how these technologies can help improve credit unions’ operations and service to members but has cautioned against an unlevel playing field. In the case of AI and machine learning, a clear regula- tory framework for testing and validating these technologies could help level the playing field with larger banks and fintech companies—benefitting credit unions and their members. AI has the ability to quickly capture and analyze large amounts of both traditional and alternative data, which can help expand access to affordable credit at credit unions. AI analysis can also produce a more robust and holistic assessment of an applicants’ creditworthiness and has been cited by regulators as a tool to expand access to low-cost, mainstream credit for millions of underserved and “credit invisible” Americans. Regulatory barriers that stand in the way of responsible AI innovation risk compro- mising the quality of member services and long-term industry viability. On the other hand, thoughtful AI adoption coupled with a flexible regulatory framework that encourages experimentation will allow credit unions to better and more cost- effectively serve present-day members and remain at the forefront of engaging unbanked and underbanked Americans. Recent Developments In July, in response to the regulators’ RFI, NAFCU submitted comments detailing how credit unions currently utilize AI, specifying that credit unions are leverag- ing AI with the goal of supporting opera- tional needs in order to deliver safe and affordable services to their members. NAFCU supports regulators allowing credit unions to create pilot programs that provide safe harbors for specific AI applications and has recommended that the NCUA form a working group of credit union industry stakeholders to pursue targeted solutions to leverage AI. NAFCU will continue to work with the reg- ulatory agencies to ensure credit unions are able to innovate and use new AI tech- nologies to better serve their members.