Early this month I moderated the panel “The Implications and Opportunities of Generative AI in FS” at Corinium’s CDAO event in Boston with David Dietrich (VP, Advanced Analytics and Governance at Fidelity Investments) and Jake Katz (Head of RMBS Research and Data Science at the London Stock Exchange Group). This was a lively discussion with a really engaged audience and it really highlighted for me the squeeze that Data and Analytics leaders are facing right now between their business leaders’ demand to hop on the GenAI train and finding a practical application for it. Data science insiders have known about GenAI for a while but the launch of ChatGPT at the end of 2022 brought awareness of it into the public consciousness, including that of senior management. Where before AI seemed ephemeral and complicated, ChatGPT made it tangible and easy. It also made AI seem a little bit like magic. As David noted, this led leaders to demand this technology, dedicating significant resources to integrate it into a broad set of applications.
But do leaders really understand how GenAI and large language models (LLMs) work and what they’re asking for?
The consensus from the audience was a resounding ‘No’. It’s tempting to shrug at this situation–this is just the latest in a long line of new technologies that seem to get everyone excited and distracted. No doubt the hype will settle down, right? Indeed, CCS Insight predicts that this is exactly what will happen in 2024 as the cost to deploy GenAI and LLMs safely and responsibly overshadows the value of the realistic applications of the technology in many situations.
Are there Generative AI Applications in Financial Services?
Does this mean that GenAI has no potential use cases in FinServ? Not at all. It’s proving its mettle with use cases in customer support, content generation, and even coming up with potential business ideas. These are all areas that offer a lot of efficiency gains and are worth exploring. But that still leaves a lot of the business that’s not currently seeing gains. And this leads me to my next point.
What’s happening to all the other data-based initiatives and AI use cases while resources are diverted to GenAI? They’re stalling, and they were already struggling (a CIO.com report says that only 53% of projects were seeing results).
I could see in the room the frustration with an audience pressured to take away their attention from problems that could be solved with applications of other, more practical forms of AI. Managing up is never easy, but I think senior leaders need to hear that GenAI, while exciting, is not the answer to every business challenge. But CDAOs have good ideas that could be valuable ideas, and it’s time to turn their attention back to solutions that make sense.
