Fintechs experiment with generative AI, but development is slow going

Businessman supervising AI
"In this last quarter, and the first innings of Q3, we are seeing more applications of [generative AI] emerge," says Pitchbook Senior Analyst Rudy Yang.
Sutthiphong - stock.adobe.com

Fintechs are increasingly considering how to integrate generative artificial intelligence into their products, especially to increase employee productivity, according to a recent report from Pitchbook.

Banks and fintechs have been using traditional AI for years in fraud detection, underwriting, compliance and other tasks, but generative AI can create new content on its own.

The number of companies and investors that have begun seriously researching, or investing in, generative AI capabilities has grown in the last few months, said Pitchbook Senior Analyst Rudy Yang. The data and analytics firm doesn't have an exact count, but more fintechs and banks clearly are starting to put the emerging technology to the test in the real world, Yang said.

However, launching finished products has been and is still a slow process, he cautioned.

"When generative AI was first emerging and making its way up through the lifecycle, there were a lot of use cases that were being talked about in the fintech universe," Yang said. "In this last quarter, and the first innings of Q3, we are seeing more applications of that emerge."

In recent months, fintechs and banks began demonstrating some success in generative AI applications, according to Yang. In July, Ramp acquired a large language model-powered customer support platform. On a virtual panel hosted by Fortune magazine this summer, PayPal CEO Dan Schulman said that the company has seen a 30% improvement in productivity where it's experimented with generative AI in code development. JPMorgan Chase recently began using large language models, the technology behind ChatGPT, to detect signs of fraud in emails.

Yang said the technology has the potential to increase operating efficiency, which is especially important now for startups eager to become profitable quicker amid a tighter funding environment. While the pace of adoption and exploration of advanced AI is picking up, Yang said financial institutions will be patient with their rollouts, due to regulatory and security concerns.

"It's going to accelerate, but it's also probably still going to be quite slow in the near term," Yang said. "Companies just want to be really diligent when they're exploring this technology. They want to make sure that they get it right. … It's very costly to even explore the technology."

Adam Shapiro, partner and co-founder at Klaros Group and regulatory advisor at Core Innovation Capital, said the demand from banks for help from fintechs that offer AI solutions is high and will continue to rise. Many banks have limited resources and expertise to build such products in-house and they are sometimes handcuffed by regulations, so they want fintechs that can deliver a compliant, efficient AI tool, he said.

"You've got to worry about what it costs to get the job done, but you've also got to worry about the standards," Shapiro said. "One of the reasons that I'm bullish on this technology is often you're making trade-offs between the two, but I think this can actually help you get both higher standards and more efficiency at the same time. And those things are pretty rare."

Although demand is high, venture capital and investment activity in fintechs has decelerated from 2022 due to macroeconomic factors. Venture capital activity in enterprise fintechs declined 69.2% from the previous year, per Pitchbook data. 

Fintech VC funding flooded into consumer-focused companies in 2021. Now, the tides are changing direction.

June 2
fundraising.jpg

Shapiro said it's still early days for generative AI. He added that there's a spectrum of perspectives among fintech investors, from focusing solely on AI investment opportunities to avoiding it entirely. 

As the amount of announcements from businesses committing to AI rises, Shapiro said that fintechs need to think about whether they want to be an "AI company" or the best in their field at leveraging AI. While AI has been a buzzword in tech, it's difficult to discern which companies are actually integrating generative AI as a component of their products, Yang said.

"A lot of funds are interested," Yang said. "I think the challenge for them is separating 'what is going to be a true use case versus the hype that we hear about.'"

For reprint and licensing requests for this article, click here.
Fintech Artificial intelligence Venture capital
MORE FROM AMERICAN BANKER