This roundtable brings design system practitioners together to share experiences and challenge assumptions about AI's role in our work.

We'll explore: What's left for design systems when AI can generate components in minutes? How do we provide AI with the right context: tokens, documentation, MCPs to produce usable output rather than slop? Why do developers still want to manually edit AI-generated code, and how do we improve adoption?

Should documentation be generated from code, or does that strip out the institutional knowledge and decision rationale that makes it valuable? Participants will leave with practical approaches from peers, a clearer sense of where AI fits into design system workflows, and where human judgment remains irreplaceable.