For years, the most frustrating part of being a designer was not having ideas. It was the distance between an idea and the moment I could finally experience it.

I could imagine a different interaction, a more useful product flow or a visual treatment that did not quite exist yet. But turning that thought into something real often required hours of production, technical help or a compromise with the tools available.

Then AI arrived, and that distance became much shorter.

I could describe a rough thought, generate possible directions, place my own design into a realistic situation, build a functioning prototype and learn from it—all before the idea had time to go cold.

That felt like magic.

It still does. But the more I use AI in my UI/UX design workflow, the less I see it as a machine that “does design.” I see it as a material: fast, flexible and sometimes surprising, but only useful when someone gives it direction.

AI can produce a screen. It cannot decide, by itself, whether that screen deserves to exist.

What AI changed for me as a designer

Before AI, making an idea visible could be expensive in time.

A designer might write a brief, collect references, sketch a flow, create wireframes, write placeholder content, design the interface and then wait for development before discovering how the experience actually felt. Every stage was useful, but the effort required to test a direction naturally limited how many directions could be explored.

AI has changed the economics of exploration.

An abstract product idea can become a rough interface quickly. A static screen can become an interactive prototype. A confusing paragraph can become five clearer alternatives. An unusual visual idea can be simulated before spending a day producing it.

This is not only my experience. Figma reports that AI product-design tools can turn abstract ideas into working prototypes in minutes, and says 85% of surveyed designers and developers believe learning to work with AI will be essential to future success.

The important part, however, is not that AI makes more things. It is that I can see an idea earlier, react to it and make a better decision sooner.

That is a meaningful change to the design process.

My practical AI design workflow

I do not press one button and ask AI to design an entire product. I use it at specific moments where speed, variation or simulation helps me think.

My process usually looks like this:

1. Understand the problem and collect context.
2. Use AI to expand the questions and possible directions.