.jpg&w=1536&q=75)
Free Lesson
Planning User Research for AI
45 min
Dec 10, 2026 12:00 PM
By continuing, you agree to Maven's Terms and Privacy Policy.
What you'll learn
Why AI breaks the 5-user rule
Learn why Jakob's "5 user" testing is now too risky for AI and how to get quantitative samples into your research plan.
Participatory AI: who decides what "good" means
Bake edge cases, trauma and harms modelling into your experience design, plan what this looks like.
The 10–30–100 ladder, from codesign to human evals
Know what each step can and can't tell you, and when to hand over to automated evals.
Why this topic matters
Most UX teams still plan AI research the way they plan website testing: five to eight users, one round, a findings deck. That worked for interfaces that behave the same way every time. AI doesn't. The same question can get a different answer, and the failures that matter most may show up in one session in twenty. It's time to update comfort levels going deeper with qual and bigger with quant sample sizes.
.jpg&w=384&q=75)
