Show Notes
"The main thing that agentic AI is going to be good at is pulling together data from multiple sources… metadata, images, device intelligence, and reconciling it to spot subtle fraud signs humans might miss.”
As fraud grows more sophisticated, financial institutions are turning to agentic AI: systems that don’t just analyse data, but act on it to detect and prevent fraudulent activity at scale.
In this episode, we explore how these AI agents are reshaping compliance and fraud prevention, and what that means for human oversight and operational efficiency.
Our special guest this week is Will Lawrence, CEO and co-founder of Greenlite AI. He shares insights on how his company’s AI agents tackle high-volume fraud, reconcile unstructured and structured data, and free investigators to focus on high-risk, complex cases.
From false positives and synthetic identities to regulatory compliance and emerging fraud vectors, this conversation highlights how businesses can adopt AI responsibly while maintaining trust and accountability in their operations.
Sumsub's website: sumsub.com Sumsub’s What The Fraud Summit: https://sumsub.com/wtf-summit/ Sumsub's LinkedIn: linkedin.com/sumsub Sumsub's Instagram: instagram.com/sumsubcom Sumsub on X: x.com/Sumsubcom Sumsub's YouTube Channel: youtube.com/@sumsubcom Thomas Taraniuk on LinkedIn: linkedin.com/in/tomtaraniuk Will Lawrence on LinkedIn: https://www.linkedin.com/in/lawrencewill Hosted on Acast. See acast.com/privacy for more information.
More from What The Fraud?
“Is anyone too smart to be scammed… or are we all just one bad moment away?”
Meet Vicky Baker, the new host of What The Fraud? from Sumsub.
“Customers don't choose between security and convenience. They expect both."
"The fraudsters have the same AI tools we do. The difference is who uses them faster."
Cryptocurrency didn't invent financial fraud, but it changed how money moves and how criminals exploit it. Now, with new US legislation on the table, the industry faces a bigger question: can a system built on decentralization ever become trusted infrastructure?