Skip to main content

GenAI: Where Should I Start?

Published Date

Corey Gross | VP, Product Management

Generative AI (GenAI) use in the financial services industry has the potential to lower costs, accelerate revenue growth, and level the playing field between larger and smaller institutions.

When we see statements like this, some may be inclined to say, “Where do I sign up?” while others may think it sounds too good to be true. The truth is likely somewhere in between.

While industry-wide research suggests that 90% of the banking industry’s working hours could be improved by GenAI tools, the best GenAI use cases and impact will be different for every organization.

With that in mind, perhaps the best response to AI’s potential is: “Where should I start?”

To help leaders frame their approach, we offer the following four recommendations.

#1: Get informed.

While AI won’t take your job, the jobs will go to those who know how to use AI. The same can be said for organizational success. Financial institutions that know how to use GenAI tools will be poised to outperform those that don’t. So, it’s time to learn.

Two perspectives we highly recommend are Accenture’s article, How banks scale generative AI for growth, and BCG’s A Generative AI Roadmap for Financial Institutions. Leaders may also want to consult Accenture’s six essentials for the adoption of GenAI.

But first, read the rest of this article. Our remaining tips will position you to help your bank or credit union make the most of what you learn.

#2: Be bold.

The adoption of Generative AI in banking will move incredibly fast, and the early adopters will benefit most from productivity gains. Early use cases include using data to personalize customer experiences through highly tailored marketing experiences and front-office interactions. GenAI is also automating the first lines of defense in fraud detection, which streamlines risk management and compliance practices.

In these and other potential use cases, try to reimagine entire processes through a GenAI lens, rather than fine-tuning one or two pieces. Tinkering around the edges won’t deliver the full potential and won’t give you a competitive edge.

#3: Be thoughtful and strategic.

However, boldness doesn’t mean throwing caution to the wind. Your strategy around Generative AI should be linked to your financial institution's values and risk strategy, and leaders should approach GenAI transformation in the same way it would consider any other type of organizational change, from a merger to a rebranding.

A governance strategy is a good starting point. It answers two key questions:

  • What policies will help us ensure that our GenAI use aligns with our values?
  • What processes will we use to ensure compliance with those policies?

There are three major regulatory frameworks that FI leaders should look to for guidance in developing a governance strategy for their organization:

The GenAI implementation process should be equally thoughtful, starting with an education and communications plan to bring all stakeholders on board and define vision and goals. As you prioritize possible use cases, organization-wide experimentation should quickly evolve into selective formal pilot programs that have executive sponsorship and agreed-upon KPIs.

#4: The importance of human insight and decision-making

People are your financial institution’s most valuable assets, and this will still be true even after you fully adopt GenAI. Organizations that focus exclusively on the tech, and forget about the people, do so at their peril.

What do we mean by this? Two things.

First, research suggests that the best use of human-machine pairing is a model where humans complement GenAI outputs rather than try to enhance them. For example, a machine-learning-based credit scoring system will make better decisions than most humans when it comes to simple credit card applications. But humans are better equipped for more nuanced situations, such as considering loans for complex financial transactions in which each application is unique.

Experts say that humans should leave the simple credit card decisions up to the bots and focus on the complicated loans, instead of thinking of the AI-driven credit card decisions as a “first pass” that needs to be human-reviewed. The second approach can significantly diminish the value of using AI.

The second takeaway is that change management is critical. As we said earlier, the best GenAI implementation plans are bold and thoughtful ones that reimagine entire processes — which are currently people-driven. The AI revolution will mean rethinking most job functions and considering how every role will interact with GenAI tools. This also leads to a reassessment of an organization’s talent needs: What AI-informed skills does the company already have, need to develop, or need to hire for? 

As your organization considers how best to integrate Generative AI tools into your business, we’re here to support you along the way. Find more resources and frameworks for each stage of your GenAI journey in our comprehensive report, “The Future of Generative AI in Service of the Banking Industry.”