Introducing UX Evals

​AI-powered products don’t behave like traditional software — but most teams are still evaluating them as if they do.

​Benchmarks and classic usability methods weren’t designed for conversational, adaptive, multi-turn experiences. Heavy reliance on these methods means that teams still don’t know whether their AI products are actually helping users.

​This live workshop introduces UX Evals — a new research method for evaluating consumer-facing AI through real user experience.

​Read more about the new methodology here.

​Who you’ll hear from:

• ​Chris Monnier, Principal UX Researcher @ Microsoft Copilot

• ​Aaron Cannon, Founder + CEO of Outset

​What you’ll learn:

• ​Why traditional UX research and AI evals fall short for AI-native products

• ​What teams are getting wrong when they equate model quality with experience quality

• ​How UX Evals capture real, first-person, multi-turn interactions

• ​Exactly how to design, run, and operationalize UX Evals on your team

​If AI behavior shapes your product experience, you can’t afford to evaluate it with outdated methods.

​Join the workshop and learn how to implement UX Evals at your org.