AI can now generate a discussion guide, synthesise forty transcripts, and draft a report in minutes. This doesn't just change our tooling — it changes the question of what makes a designer or researcher good at their job.
As a Music Psychologist and UX Researcher, I've come to believe the answer isn't in the tools at all. It’s in a skill musicians have trained for centuries: improvisation. A jazz solo looks like instinct alone, but it's actually thousands of hours of practice, compressed into a few live seconds, played inside changes someone else wrote. Designers and researchers do this every day: the researcher who abandons the script to chase an unexpected answer, the designer who knows a layout is wrong before they can say why.
Now AI is writing some of those changes too. The real question is what it takes to keep improvising when the changes keep shifting. The instinct still has to be ours.
This talk draws on research into AI self-efficacy, anxiety, and psychological safety to reframe our evolving roles as designers and researchers: less rigid methodology executed step by step, more jazz musician reading the room, knowing which “wrong note” is actually the most interesting finding. You'll leave with a new way to think about rigour, listening, and error — and why our edge was never speed of output, but knowing which chord you're playing over.
