I’m a fan of Figma, but I’m a little nervous for their future. If their AI workflow isn’t at least as good as Claude’s ability to generate and iterate on an actual react prototype I’m afraid they may become irrelevant fairly quickly. I get not everyone understands how to code, but the lines between engineer and designer are exponentially becoming blurred. I’m surprised their acquisition of Diagram hasn’t yielded anything interesting at all yet.

Great question! I think the conversation around Figma AI being "overhyped and underdelivered" really depends on user expectations.

Figma’s recent AI features—like auto-generating UI layouts or content suggestions—definitely show promise. But as with many tools integrating AI, the results can feel a bit shallow if you're expecting fully polished designs. It seems more like a productivity assistant than a creative replacement. That said, it’s still early days, and AI in design is evolving fast.

The key issue here might be implementation. Tools like Figma are great canvases, but the real magic happens when companies bring in a skilled AI developer who can fine-tune models and integrate AI in ways that are more tailored to a specific workflow.

So rather than relying solely on built-in AI, teams might benefit more by developing custom AI tools or plugins that match their design process. That’s where the real potential lies.

Curious to hear others' experiences—have any of you tried customizing Figma with AI-powered plugins or automations?