Why generative AI projects fail in highly regulated industries — and how yours can succeed
Get the latest in your inbox
With nearly 20 years in financial services and technology experience, Matt offers valuable expertise for organizations in regulated industries that are interested in making the most of innovation.
It seems Generative Artificial Intelligence (GenAI) is everywhere. Businesses in industries ranging from pharmaceuticals to waste management are taking advantage of the technology to accelerate growth – with good reason. Automating mundane tasks frees valuable resources for more important activities. Pulling smart insights from complex data can quickly lead to increased revenue and better business outcomes. Making expansive knowledge accessible to the entire business helps everyone become more efficient and effective.
Exercising (Extreme) Caution
Unfortunately, the financial services industry has been slow to adopt GenAI. In fact, as I noted in this recent article for American Banker, a full 30 percent of financial services organizations ban the use of the technology within the company altogether. While there are multiple reasons for their apprehension, the fear of inaccuracies, security concerns, and regulatory scrutiny top the list.
Regulated industries like financial services face unique challenges when adopting any new innovation. Penalties and fines are prevalent, making organizations highly risk-averse to emerging technologies. And, because the level of accuracy required is higher than in non-regulated industries, data quality must be at the forefront of any GenAI project. At the same time, many of the GenAI tools currently available produce results that are unexplainable, making concerns about regulatory inquiries widespread – and warranted.
Why AI Adventures Go Wrong
The GenAI projects that financial services organizations undertake often fail for the same reasons they do in other industries. Rather than identifying specific areas where the technology can truly add value, it’s common for teams to simply get excited about trying something new and attempt to do too much, too soon.
Projects are also frequently launched without a focused goal and well defined scope. The lack of a clear aim to guide the effort often leads to failure and wastes months of effort (and a substantial financial investment).
Successfully deploying GenAI is a complex undertaking, requiring a measured approach and the right expertise and resources.
How to Succeed in GenAI by Really Trying
The good news is that despite the challenges, financial services organizations aren’t doomed to fail with GenAI. A cautious, thorough, and well-planned effort can lead to substantial business benefits. Here are a few best practices to keep in mind:
- Start small. Identify opportunities to deploy GenAI where the bar to entry is lower.
- Find a champion. Tap an influential colleague who is positive about GenAI capabilities, and can advocate for the effort across the organization.
- Define the scope. Clearly state the project’s range and goals, and provide well-defined, measurable business outcomes.
- Make it transparent. Ensure everyone understands how the solution should work - how it receives inputs, combines them, and develops results. (Sufficient transparency is essential in the event outputs need to be explained to regulators.)
- Keep security paramount. Implement solutions with proper safeguards to ease leadership concerns and demonstrate to regulatory authorities a secure, responsible approach.
- Make results clear. Root design in explainable results, and ensure full transparency of how models are used and how data is accessed.
- Bring the team along. Train internal partners as the project progresses to accelerate adoption and reduce costs over the long run.
GenAI can be an invaluable tool for financial services organizations. The key is to begin by focusing on solvable challenges, build on the momentum of smaller successes, and work your way toward more complex efforts as comfort levels rise.
The Company You Keep
Finding the right partners can have a massive influence on the success of a GenAI project. Choosing teams and vendors that bring established expertise to the effort can point you in the right direction from the start. Technologies that provide proper guardrails and clear explanations of results can do much of the heavy lifting.
The Right Tool Can Unlock Success
The benefits of many existing GenAI solutions can be limited by several factors, including costs, complexity, and weighty training requirements. Time-to-market can be extremely lengthy, and even then, the results can be opaque, cryptic, or inaccurate. That’s why many financial services organizations are turning to Stardog Voicebox.
Voicebox is the world’s first AI Data Assistant that allows anyone to ask any question of their enterprise data, and get a fast, secure answer that is hallucination-free—and fully explainable. This transformative new tool makes it easy to empower everyone in the organization with the data they need to make smart decisions, without the need to acquire new skills.
Moving Forward the Right Way
GenAI is changing how business is done, everywhere. While regulated industries have been slow to adopt GenAI—for good reasons—the challenges they face are not insurmountable. A clear focus and an approach that ensures security and transparency will go a long way toward achieving business results previously out of reach. Rather than increasing regulatory scrutiny, demonstrating a cautious approach and attention to detail and providing fully explainable results can actually improve your business standing with authorities. And that can pay off well into the future.
To learn more about GenAI in financial services, join me for an upcoming webinar: The AI-Powered, Data-Intelligent Wealth Advisor, May 22 at 12:00 ET.
