Many banks are doing an excellent job at embracing generative AI (Gen AI). However, not all have a specific Gen AI strategy, and they should make their journey more efficient. A quality Gen AI strategy aligns an organization’s objectives, its use cases, and the delivery mechanism and metrics that measure business value. In this article, we highlight some Gen AI strategy insights for community banks and provide tools to help bankers advance their programs.

Gen AI Strategy Checklist

Below is our starting checklist of potential elements for consideration in order to evolve a quality Gen AI strategy document. The overarching goal of the document is to align the organization, so each business line does not go off in different directions sourcing their own Gen AI tools. For example, in the next year, does the bank want to focus on making its employees more productive or enhancing customer experience. Few community banks have the resources to accomplish both, and both are sizeable efforts. Meanwhile, hundreds of bank vendors are coming out with their own Gen AI tool, many of which are distractions and not worth the risk of rolling out.

Gen AI Strategy Checklist

Goals and Use Cases of a Gen AI Strategy

Many banks started off trying to manage their Gen AI strategy and governance by application. This turned out to be a bad idea as applications are not specific enough and one application might be good for one use case and not for others. In 2025, banks evolved in managing their goals and objectives through use cases. That is, they adopt the existing business strategy of the bank and focus Gen AI on use cases that help achieve the bank’s goals.

By far, the most popular use case is enabling employees to access a bank’s knowledge base more efficiently, such as its policies, procedures and supporting product documents. This use case not only has a direct impact on employee time savings and an easily calculated return on investment, but it allows employees to get comfortable with the technology. Another popular use case is document intelligence, which we went wrote about in-depth HERE.

Gen AI Strategy Use Cases

Banks can also take their use cases and rank them according to business value and feasibility like we did with a set of use cases using the Gartner model. This helps provide a level of specificity and priority ranking based on a metric such as time-on-task, return on investment, or time saved.

How Will The Bank’s Gen AI Strategy Support Agentic AI?

Last year, banks were just learning about AI Agents (HERE). This year, banks are starting to think about how intelligent Gen AI-driven agents will be deployed. Applications like Salesforce, Microsoft and ServiceNow have combined Gen AI and agent offerings forcing banks to have their Gen AI and agentic AI strategy as one. Any bank writing a Gen AI strategic plan now should consider including their goals and roadmap for agentic AI.

Will The Bank Have a Core Gen AI Model?

One of the main questions to answer is should a bank have its own Gen AI model in which to use for a variety of use cases and deploy across the organization? This model could be its own model such as our Tate (HERE), Microsoft CoPilot, or a third-party stand alone model such as a Posh. A bank will have thousands of its own use cases where it can use its model and create its own solution to satisfy its use cases, or plug its model into third-party applications to do the same. The advantage here is that a bank’s primary LLM would be flexible, agile and a known risk instead of using a variety of third-party models.

Will the Bank Rely on Various Vendors for Its Gen AI?

Smaller banks may want to pursue a strategy where it just uses third-party models. While harder to govern, deployment might be easier. A bank would likely lose flexibility, drive-up cost, and have a variety of user experiences, however, would gain the advantage of being able to deploy multiple models across different uses cases quickly.

Governance and Data

A Gen AI strategy is like any other strategy with the exception that it is a brand-new field. Because Gen AI enterprise use is in the nascent stages, a workable strategy should highlight how Gen AI will be governed, monitored, and managed. In addition, a Gen AI strategy should also be interconnected with any data/analytics strategy since many Gen AI applications will be powered by the bank’s data. Further to this point, a Gen AI strategy should connect to a bank’s risk model management effort, should it have one.

See A Sample Set of Use Cases

Within an Gen AI strategy, banks should consider providing use cases that have a strategic frame around them complete with how the use case aligns with business goals, the use case’s priority, the vendor if identified, the goals of the use case and the expected results complete with quantifiable metrics if possible.

To download a set of sample use cases written within a strategic framework, go Sample Gen AI Strategy Use Cases 020425v.

Coming Up Next

Banking leaders can leverage Gen AI to address key business and operational challenges. Bank technologies and Gen AI supporters should continue to evolve their Gen AI strategy to learn how to more efficiently and effectively deploy Gen AI to shape their roadmap.

Soon, we will dive deeper into how a bank’s Gen AI strategy is interconnected to its business, data, and model management strategy, discuss how to risk score use cases in order to test for materiality, highlight testing requirements and go deeper into managing the Gen AI inventory.

 

Tags: , , , , Published: 02/04/25 by Chris Nichols