2026: The AI I’m using in my UX and content work

The AI I'm Using in My UX Work — 2026 Update

It's 2026, and AI has become a genuine part of my daily workflow — not as a novelty, but as a set of tools I reach for the way I used to reach for a reference book or a sticky note.

I use Claude for longer-form strategic thinking: drafting stakeholder documents, working through information architecture problems, developing content systems. ChatGPT handles the faster lookups — Figma Make prompts, quick research questions, anything where I need a reasonably good answer in thirty seconds rather than a considered one in five minutes. They're complementary in practice in a way I didn't fully expect when I started using both.

Figma Make has been the biggest shift in how I work. Over the past few weeks I've built four platform site prototypes with it, and the speed-to-fidelity ratio is genuinely different from anything I've used before. It doesn't replace the thinking — the IA, the content strategy, the decisions about what a screen needs to do — but it compresses the time between idea and artifact in a way that changes how I iterate.

Relume has been helpful oner the past two years for content mapping and quick ideas that clients can grasp but with the way Figma is moving I’m not sure I will use Relume as much unless I dig deeper into Webflow.

Midjourney and OpenArt have been invaluable for a current project involving a digital library — specifically for storyboarding and visual concepting around complex, esoteric subject matter that doesn't have obvious stock photo equivalents. The ability to generate reference imagery for animation and illustration briefs has saved significant time at the visual development stage.

Grammarly I use constantly, mostly for the speed of it. When you type fast, you need something watching the back end.

Relume has become part of how I scaffold early-stage website and component work — particularly useful for generating sitemaps and wireframe structures quickly before moving into higher fidelity.

All in I'm spending $70+ a month on AI tools. That's a line item I didn't have two years ago, and it's one I don't question.

Looking back: what I wrote about AI in 2023

It's useful to look at where I was three years ago. In 2023 my toolkit was: Fathom and Otter for meeting transcription, Dall-E and Midjourney for visual exploration, Wordtune for writing support, and ChatGPT 4 for ideas and questions. I was watching Uizard and Galileo with interest but not yet finding them useful enough to integrate.

What's changed is less about the tools themselves and more about the depth of integration. In 2023 I was experimenting. In 2026 I'm relying — and knowing which tool to rely on for which kind of problem is now a skill in its own right.

For staying current I follow the Superhuman AI newsletter and Jakob Nielsen's substack. Nielsen has been writing about UX longer than most tools in this list have existed, and he brings a useful skepticism to AI hype that I find clarifying.

Previous
Previous

Prompt precision: 60% increase in quality and accuracy if you talk to GPT the right way

Next
Next

Responsive design beyond breakpoints with Figma’s variables and modes