Prof. Rajeeva Karandikar

From classroom assignments to industrial software engineering, Generative AI has moved from a research curiosity to a ubiquitous utility. While tools like Gemini, ChatGPT, Grok and Claude offer "striking confidence" in their outputs, they often operate without a baseline of logical truth. This talk explores the vital intersection of the available data and computation power with human intuition, domain expertise, and mathematical discipline. Drawing on illustrative examples from the past 150 years, we examine why "Big Data" and high-performance computing are insufficient for high-stakes decision-making. We will discuss how the mathematical sciences serve as the "ground truth" for interpreting AI-generated content and assessing risk. By highlighting the hidden dangers of purely data-driven conclusions, this talk reinforces why the next generation of thinkers must be grounded in the foundational principles of Statistics, Probability, and Modeling to truly master the machines they use.