For financial services, 4 keys to operationalizing gen AI

Two years after the technology burst onto the scene, adoption remains decidedly mixed. Here’s why—and how banks can course-correct.

With its promise of innovation and net new revenues, generative AI initially seemed like a banking paradise. Instead, early use cases have emphasized productivity—and fallen far short of the promised land. So far, too few proofs of concept make it into production; “PoC fatigue" has set in. 

How do banks course-correct and pivot to gen AI that boosts the bottom line?

The answer is deceptively simple: focus. Amid the frenzy, maintaining clear direction is key to operationalizing gen AI and rolling out a consistent flow of high impact use cases backed by a system that makes them happen. This targeted approach balances the risks of gen AI with the technology’s potential for innovation. It starts with data and then creates an enabling infrastructure and governance organized for the long term. Most important, it never loses the human element.

Facing the hurdles

Despite banking leaders’ optimism about gen AI’s revenue opportunities, most outcomes to date have been underwhelming. While it’s good news that 63% of financial services companies have moved gen AI use cases into production, according to a recent survey, it’s sobering that 35% are still merely evaluating or testing use cases. Equally worrisome, of those who are using gen AI in production, just over half report increased revenue as a result.

Additionally, banks and financial services firms face a range of gen AI-related challenges, from inadequate technology infrastructures to talent shortages. IT and business leaders alike worry about product stability given the frequent release of gen AI product updates. In the US, where federal regulations have yet to be issued, banking leaders have expressed concern about safe, secure use of the technology.

While the financial services sector faces unique constraints, it’s not alone in its gen AI challenges: Gartner predicts at least 30% of gen AI projects will be dropped after the PoC stage by the end of 2025 due to poor data quality, inadequate risk controls, escalating costs, or unclear business value.

Operationalizing gen AI: How to course-correct

While models come and go, gen AI only makes sense when you start with outcomes and relevant data. The early focus on use cases and experimentation with PoCs is shifting as banks struggle with how best to put gen AI into production and extract business value. 

There are four keys for banks seeking to operationalize the technology and pave the way for a continual flow of high-impact use cases:   

1. Start with data. When it comes to the availability of data, banks have made great strides in everything from reporting and governance to more informed decision-making. Achieving data maturity, however, remains a struggle—and the lack of progress here is an impediment to gen AI adoption. In one survey, only 9% of banks were categorized as “data-first organizations” fully embracing a data-driven mindset and using data for nearly every decision.
   
   That survey also found that while many financial services organizations (39%) were in the “innovation-ready” stage of digital advancement, 14% of institutions are slower in adopting advanced capabilities and still rely heavily on third-party digital providers for data. The upshot? Building a data foundation is the first step to operationalizing gen AI. It ensures your business eliminates silos, ensures quality, and creates discipline across the organization.
   
   
2. Create an enabling infrastructure. It takes a range of foundational capabilities to scale gen AI adoption across the enterprise. Large language model (LLM) adapters, connectors and prompts are a good start. Ally Financial provides a forward-thinking example of the high-impact uses cases that can benefit banks’ bottom line: the company has rolled out a cloud-based platform that over time will provide access to multiple LLMs, providing users with the flexibility to query the model of their choosing. 
   
   
3. Set up governance that’s organized for the long term. Governance is about tracking and measuring use cases to better understand project outcomes. But it’s also about developing support functions that ensure your organization is ready to manage—and adapt to—the accelerating speed of change. For example, we partnered with a payment card provider that co-invested in a shared AI/machine learning innovation hub and is already reaping the benefits. The hub’s approach to design thinking provides the company with a framework to prioritize and implement use cases. 
   
   
4. Keep it human. Research by Cognizant and Oxford Economics found generative AI has the potential to disrupt 90% of jobs over the next 10 years. In financial services, we can expect to see impact on tellers, traders, and advisors, to name just a few roles. However, the goal of gen AI is to augment human capabilities, not replace them. Emphasizing the human aspects of gen AI allows a far more nuanced approach to enterprise change than many headlines would have us believe. Take software code development, for example. For experienced, full-stack, high-end developers, gen AI tools may offer modest to no benefits. But they could be enormously beneficial for entry-level developers just beginning their careers and still finding guidance helpful for tweaking code and writing syntax.

Financial services leaders need to both focus and think bigger. Even though we’re in the early days of gen AI, a down-to-earth, pragmatic approach will get us on the road to real outcomes.