Why Better Data Beats Bigger AI Models: Enterprise AI Executive Rohit Chauhan
CDO Magazine Podcast SeriesShow Notes
Rohit Chauhan, former Executive Vice President of Artificial Intelligence & Fraud Solutions at Mastercard, tells Robert Lutton, Vice President of Sandhill Consultants, why a more powerful AI model will not fix fraud detection on its own. Chauhan argues that trust in AI comes from the data architecture underneath it, not from the model's output, and explains how connected context can flip a flagged transaction into a normal one.
More from CDO Magazine Podcast Series
Shebani Baweja, Group Chief Data Officer at Standard Chartered, speaks with Bryan Lee, Group Head of Data at United Overseas Bank, about why governance, agility and innovation reinforce each other when they are designed well. Baweja also maps how the CDO role is moving from risk and control leader to strategic enabler as AI becomes embedded across the bank.
Beth Hiatt explains why AI initiatives need clear business priorities and why governance programs can lose traction when organizational maturity is overlooked. Jessie Smith, Chief Product Officer at Ataccama, explores how data leaders can assess readiness, adjust execution, and connect AI use cases to measurable business outcomes.
Legacy technology architectures were not designed for artificial intelligence, making tail-end data reviews a major bottleneck for scaling enterprise AI. In this interview, Shebani Baweja, Group Chief Data Officer at Standard Chartered, speaks with Bryan Lee, Group Head of Data at United Overseas Bank, about embedding data controls directly into development pipelines. Discover how automating data governance and building trusted data products accelerates enterprise innovation.
How can government agencies prepare data for advanced analytics and AI while maintaining quality, governance, privacy, and public trust? Neil Cooke, Chief Data Officer at the State of Texas, discusses the challenge with Dr. Adita Karkera, Chief Data Officer, Government and Public at Deloitte, including agency data maturity assessments, responsible AI practices, and practical data literacy.
Kyle DuPont, Founder and CEO of DataXray, speaks with CDO Magazine Editorial Board Member Maribeth Achterberg, that most enterprise AI strategies stop at a stack of Copilot licenses with no way to measure return. DuPont breaks down why a typical enterprise project takes roughly eighteen months to move from proposal to a budgeted rollout, and argues that real safeguards start with knowing exactly where sensitive data, especially within unstructured files, lives.