What separates people who achieve remarkable results with AI from those who struggle to get reliable, accurate, or useful outcomes from the exact same tools?

In this episode of Product Rising, host Shannon Peavey sits down with Stanford researcher and BigSpin AI founder Moritz Sudhof to explore one of the most overlooked challenges in AI adoption: the human factor.

Drawing on research conducted with Stanford professor and Bigspin co-founder Chris Potts, Moritz shares surprising findings from the analysis of tens of thousands of real-world AI conversations. Their work reveals that the difference between successful and unsuccessful AI use often has less to do with the model itself and more to do with how people interact with it.

Why do expert users encounter more AI failures than novices, yet achieve dramatically better outcomes? Why do so many users unknowingly accept flawed outputs? And what skills, behaviors, and product design choices can help close the growing divide between AI power users and everyone else?

This conversation explores the emerging field of AI fluency, the risks of treating AI like an oracle or a vending machine, and why the future of successful AI products depends on designing not just the model, but the interaction between humans and machines.

Moritz is part of our exclusive Product Rising series on AI Ethics, Safety & Responsibility, where host Shannon Peavey speaks with researchers, builders, policymakers, and practitioners working to shape a future where AI delivers meaningful value while preserving human judgment, agency, and trust.

Whether you're a product leader, founder, builder, manager, or everyday AI user, this episode offers practical insights into how to get better outcomes from AI and why your own behavior may be one of the most important variables in the equation.

CHAPTERS

00:00 Why Some People Get Better Results from AI

01:22 Moritz's Journey: From Language Research to AI Products

03:42 AI Is Not a Human Replacement

06:49 The Coaching Experiment That Changed Everything

09:14 Same AI, Different Outcomes

10:43 Designing User Behavior, Not Just AI Behavior

12:13 The Research Behind AI Fluency

15:14 The User Fluency Paradox

17:01 Why Expert Users See More Failures

19:18 Invisible Failures and Silent Mistakes

21:12 The Skills Every AI User Needs