Log in  |  Forgot Password?

Your LLM has opinions it doesn't know it has

by John Haggerty | at Minnebar20

I've spent the last year running thousands of behavioral tests against large language models, and the patterns are consistent and unsettling. LLMs anchor heavily on early information. They escalate commitment to bad decisions. They treat vivid examples as more statistically meaningful than they are. If you're building products on top of LLMs, these aren't edge cases. They're systematic tendencies that show up in production. This session walks through what cognitive bias in AI actually looks like in practice, how you can probe for it in your own systems, and what to do when you find it. Come with questions about whatever you're building.

John Haggerty

John Haggerty is the founder and CEO of BiasHawk, a cognitive bias-monitoring platform for LLM-powered AI systems, and the founder of The PM Insider, where he serves as a product coach. He is an adjunct professor at Case Western Reserve University, where he teaches an MBA course on Building AI Products covering AI fundamentals, LLMs, product development, and the technical, organizational, and societal risks of deploying AI systems.

John has spent 25+ years in product leadership at companies including Datasite, Prodege, Highway.AI, and others. His work sits at the intersection of behavioral psychology, AI systems, and product decision-making.

He is based in South Minneapolis.

Links: - BiasHawk - The PM Insider - LinkedIn

Are you interested in this session?

This will add your name to the list of interested participants. It will help us gauge interest for scheduling purposes.

Interested Participants

Similar Sessions

Does this session sound interesting? You may also like these: