When it comes to research, AI is good at making you average

"AI is really good at making you average."

This seemed to resonate with folks in my presentation yesterday on how researchers can "coexist" with AI.

Thanks again to Jemma Ahmed and Rosenfeld Media for inviting me to speak on AI at Advancing Research 2025.

Day 1 highlighted the many ways the research profession is changing, not the least of which is AI. As more designers, PMs, founders, and researchers themselves begin to use generative AI research tools for research synthesis, this has many researchers wondering whether it's good enough to replace their work.

The full answer is nuanced (I have the lit review to prove it), but the short version is it's a lot like using AI for anything else: AI output is "adequate." If you're not a researcher or don't have much experience, it will probably make your research synthesis better. If you're a top-performing researcher, it'll likely make you a little worse. (And worse still if you're doing highly innovative work or designing for niche audiences.)

Whether it makes sense to use AI for synthesis will depend on the specifics of your situation, your experience level, your AI tools, and your data.

I'll be doing a workshop later this month to explore the pros and cons of AI for UXR. To hear more about that (and upcoming AI office hours), you can sign up for updates here.