Almost overnight, AI, specifically Generative AI (or GenAI), has become the hot button du jour. Virtually every company I review has decided there is some AI component that must be a part of their service. Setting aside the fact that shoehorning AI into an unwarranted solution doesn’t always make it better (and in fact could make it worse …), there is no question that the rapid advancement of GenAI solutions with improved capabilities making results more contextual, accurate and less likely to hallucinate make the intelligent use of GenAI in financial services a strategic imperative. It’s a powerful tool and bankers may ignore it at their peril. Yet, there are legitimate concerns about using GenAI in banking applications. Especially as it relates to making decisions that affect customers or giving customers access to services powered by GenAI as opposed to experienced bank personnel. There are issues of how PII (Personally Identifiable Information) data will be protected, ethical questions about the output created by GenAI and the need to ensure that all output is accurate and free from gross error. All of these concerns can be remediated with the proper approach to how GenAI is deployed, managed and monitored. I am guessing that your institution has had discussions about the use of GenAI and for many of you the decision was made to limit or refuse access at the corporate IT level. Suffice it to say, you likely have some internal skeptics and the question is how do you get them over the hump of wanting to avoid any use of GenAI. I came across a great article from The Financial Brand focused on this very issue. The title is “How to Get Your Remaining AI Skeptics Onboard.” You can access full article here. The essence of the article focuses on the strategies you would use to convince those not enamored with GenAI to understand the strategic nature of its use. The three key elements needed are 1) Address root fears about AI, 2) Start small but get started now and 3) Make compliance an AI ally. Let’s examine each of these elements in more detail: Address the AI Fear Factor … with a Compelling Narrative – The author of the article rightly points out that many of those who hesitate to embrace GenAI have a limited understanding of it or experience with it. The applications for financial services are wide-ranging, but if the immediate focus is on customer-facing applications, then that could cause legitimate concern. There is not enough confidence yet for an AI powered chat to ensure that the information provide by AI is accurate, contextual and non-hallucinatory. Focus instead on using GenAI for text and image generation that powers marketing and social media campaigns, better email communication and inter-departmental conversation/idea starters. Also consider the use of AI in closed, Small Language Model (SLM) applications like uploading all loan ops related documents and having it become the repository for answering questions. The SLM ensures that only the documents you provide are used for answers, eliminating hallucinations. GenAI is amazing at detecting anomalies and the applications for fraud avoidance are huge. The key is to make sure that there is a significant upside displayed for using AI to eliminate wasted time and avoid losses without any customer-facing issues, which would make use of AI less scary. Start Small But Start Now – I cannot emphasize enough the need for bankers to not “stick their heads in the sand” regarding use of AI. The problem for many senior leaders is considering the deployment of AI to be enterprise-wide: a huge project that would require extensive planning and resources. The better strategy is to start small. When I conduct innovation workshops at institutions who are interested in expanding employees’ focus on innovation, I urge them to start small. Small innovations are easier to envision and deploy, but over time, small innovations lead to bigger innovations as people and processes become more familiar with making creative ideas turn into meaningful innovations. As the author of the article points out, starting small also allows the institution to learn from early mistakes and adjust the process to ensure that the most value from innovations can occur in the long-term. Establish a small but cross-functional group of employees who can begin using GenAI and have them report on how they are using it. Make AI a Compliance Ally – Risk avoidance is endemic to banking, and GenAI initiatives can be viewed as increasing risk as they are deployed. As bankers, we are experts at identifying and managing risk. It’s what we do. Rather than allowing concern over risk ground us from action, we should be proactive in identifying GenAI risks and establish the policies and procedures to remediate those risks. Specifically address how your use of GenAI will detect bias and discriminatory outcomes to ensure ethical use. Consider this statement from the article, “AI isn’t just about automation, it’s about trust. Institutions that integrate AI with clear goals, ethical safeguards, and strong data practices will not only comply with regulations but gain a competitive advantage.” Exactly so. I decided to ask ChatGPT how it would address AI skeptics. My query was, “Briefly summarize ways to change the mind of an GenAI skeptic at a financial institution” – Chat GPT generated the following: Here are several effective ways to influence a GenAI skeptic at a financial institution, especially in a brief, executive setting: Lead with Risk Reduction and Efficiency Frame GenAI as a control tool, not just an innovation tool — e.g., use it to detect anomalies, automate compliance checks, or reduce human error. Show audit trail transparency in GenAI decisions (e.g., using RAG systems or explainable AI). Highlight Secure, Use-Specific Wins Share low-risk, high-impact use cases like: Automating internal documentation. Enhancing fraud detection with pattern analysis. Accelerating customer onboarding with AI-powered document handling. Showcase Industry Momentum Point to regulatory-friendly adoption (e.g., JPMorgan’s IndexGPT, or Morgan Stanley’s use of OpenAI for advisor support). Use peer benchmarks: “Your competitors are already using this for ___.” Demonstrate Human-Centered Collaboration Reposition GenAI as a copilot, not a replacement. Use examples of augmented decision-making, not autonomous systems. Pilot First, Expand Later Propose a sandbox or controlled pilot in a non-sensitive area (e.g., marketing copy, RFP response generation). Emphasize measurable outcomes: time saved, errors reduced, insights uncovered. GenAI is powerful and yet still in its infancy. We cannot even fathom the capabilities it will have in the coming years. However, I truly believe that experimenting with it now and understanding its myriad of benefits and potential pitfalls will enable us to truly be in a position to most benefit from the future opportunities that GenAI will create. To paraphrase an ancient Chinese proverb, the best time to start using AI was yesterday, the next best time is today. Get on it! The views expressed in this blog are for informational purposes. All information shared should be independently evaluated as to its applicability or efficacy. FNBB does not endorse, recommend or promote any specific service or company that may be named or implied in any blog post.