EXL integrates Nvidia's transaction model as revenue climbs 14%

EXL is embedding Nvidia's transaction foundation model into its AI platform, giving banks a unified intelligence layer for fraud, risk and personalization.

EXL is integrating Nvidia's Build Your Own Transaction Foundation Model into its AI platform, letting financial institutions train unified models on billions of transaction events for fraud detection, risk management and personalization. The New York-based data and AI company, which employs about 67,000 people globally, embeds the Nvidia developer example into its EXLerate.ai platform.

"Every financial institution is sitting on decades of transaction data that holds the answer to their biggest challenges — fraud, risk, underwriting, customer experience," Kevin Levitt, senior director of global business development for the financial services industry at Nvidia, said. "By integrating the Build Your Own Transaction Foundation Model developer example with EXL's deep financial services expertise, financial institutions create a shared intelligence fabric that drives smarter decisions and better customer outcomes."

The integration allows banks, payments firms and insurers to train transformer-based models on proprietary datasets including payments, transfers and customer behavior signals, replacing fragmented, rules-based systems with adaptive transaction monitoring. EXL reported $570.4 million in first-quarter revenue, up 13.8% from a year earlier, according to its latest earnings.

"The next frontier of enterprise AI is built on unified transaction intelligence," Vikas Sharma, head of banking and capital markets at EXL, said. "For years, firms have built separate models for fraud, underwriting, risk and customer engagement, but the future belongs to institutions that can create a unified intelligence layer across all transaction activity."

EXL shares trade at $29.43, about 39% below their 52-week high of $37, while the median analyst price target across six analysts stands at $44, implying roughly 50% upside. JPMorgan's Puneet Jain rates the stock at $43, while Stifel's David Grossman has a $46 target. The partnership expands EXL's existing collaboration with Nvidia and positions it to capture more AI infrastructure spending from financial institutions moving beyond siloed legacy systems. For Nvidia, the deal extends its reach into financial services transaction intelligence, a market where banks have historically relied on custom-built rules engines rather than foundation models.

This article is for informational purposes only and does not constitute investment advice.