How Generative AI Changed My Design Process (Without Replacing Design Thinking)

I’ll be honest! When generative AI started making headlines, my first reaction wasn’t excitement. It was scepticism. I didn’t even give it a try for months. I was so convinced that this wasn’t going to stay.

As an experience designer, I’ve spent years learning how to listen to users, synthesise messy research, connect the dots between business goals and human needs, and translate all of that into experiences that actually make sense. Could a tool really do any of that? Doesn’t it require human empathy to understand this fully?

The short answer: No. But it could do something else entirely, and that changed everything.

Where I Was Losing Time (Without Realising It)

Design thinking is rarely the bottleneck. The bottleneck is everything around it.

Summarising hours of user interviews. Building slide decks at midnight. Writing research briefs. Structuring competitive audits. Formatting workshop templates. These tasks aren’t where great design happens, but they were consuming a significant chunk of my week.

That’s where Generative AI stepped in.

What AI actually helps me with

Here’s where AI has genuinely transformed my workflow as an experience designer at Adobe:

Research Synthesis: I feed in raw interview notes, and AI helps me identify patterns, surface recurring pain points, and draft initial insight statements. I still validate and reframe everything. But the first pass happens in minutes instead of hours.

Planning & Facilitation: Writing workshop agendas, creating research discussion guides, and structuring sprint plans. AI cuts the prep time dramatically. I describe the goal, the audience, and the constraints, and I get a solid starting point instantly.

Presentations: Translating design rationale into executive-ready narratives used to take me forever. Now I use AI to structure the story arc, then I layer in the design decisions, data, and nuance.

Ideation Support: When I’m stuck, AI is a great thinking partner. I’ll describe a design problem and ask it to challenge my assumptions or generate alternative framings. It’s like rubber duck debugging, but the duck talks back.

Where AI Still Falls Short

AI doesn’t understand why a user hesitated before clicking. It doesn’t make sense when a stakeholder nods but isn't convinced. It can’t read a room. It can’t weigh the cultural context that wasn’t in the training data. And it absolutely cannot make the judgement call that turns a good experience into a memorable one.

Design thinking, the real kind, requires empathy, intuition, and the ability to sit with ambiguity. That’s still deeply human.

The Shift in My Perspective

I stopped thinking of AI as a threat the moment I started treating it as infrastructure, like having a faster laptop or a better prototyping tool. It doesn’t replace what I do. It removes the friction around what I do.

And that’s actually the point.

The designers who will thrive aren’t the ones who resist AI, nor the ones who blindly rely on it. They’re the ones who know exactly where human judgement is irreplaceable and use AI to protect that space.

I used to spend energy on tasks that didn’t require a designer. Now I spend it on the ones that do.

That’s the real shift.