Are UX Designers still relevant in a world with AI?
Learn what it means to be a designer in a world surrounded by AI, and how you can bring continued value to your stakeholders.
Many companies are utilizing AI now, in fact, if they are not, they are probably going to fall behind fairly quickly. But does this usage of AI always mean replacing the people who worked those jobs? Not necessarily. In fact, I would argue that the companies that know what it means to adopt AI will absolutely keep their team onboard.
In fact, as fresh as AI is, I've already seen many products designed entirely by AI. Some at startups, some at enterprise level corporations. They all feel similar. Icons next to titles, cards everywhere, and the navigation or purpose of the product never really refined. It's as if the person who requested, simply gave a generic prompt and ran with it.
Let me give you a concrete example. At a financial institute I was working with, they had built a dashboard for internal use entirely by AI. (Actually they paid thousands to an agency who did this for them...a cautionary tale to tell another time). The purpose of it was to give insights into team and product performances, from time per phone call to the amount of accounts opened. The data compared this month with past months to find trends.
The problem they were facing was usage and value delivered. The very first question I asked was: what do the customers really want to get out of this product? Because currently, 90% of the data on this page is past data, which means 90% of this page is static, unchanging, and therefore no reason to keep revisiting. The most valuable time is at the end of each month or every quarter, depending on where you are on the org chart.
As we dug deeper, we found more elementary UX issues, such as data that should be together for a full picture scattered across different pages. This meant users were opening multiple tabs or splitting windows to compare data.
I wish this was an extreme scenario, but it really is not. Most products that are built entirely with AI by people who haven't had the training to understand product architecture have the same issues.
What does this mean for those in the design field?
Let's first and foremost be honest about what's actually happening, because pretending nothing is changing doesn't serve anyone. Some roles are shrinking. Some are being restructured. If your entire value was "I can execute a Figma file from a spec," that work is genuinely getting cheaper and faster to produce with AI and companies are act accordingly.
But here's what's not getting automated, and it's the same thing that solved the dashboard problem above: systems thinking. AI does exactly what you want, it's not going to question on a deeper level why you asked for a specific feature. It may be able to generate a hundred variations of a login flow but it cannot tell you that your internal dashboard is failing because nobody defined what "usage" should mean. AI is like having an intern to execute, not strategize.
So how does a designer stay ahead? The tools and way you work must continuously evolve. Here are a few shifts I'd like to point out specifically.
The requirements for hiring has changed
Job postings are starting to reward people who can direct and evaluate AI output to bring quality solutions faster — not just produce it. Prompt fluency is becoming table stakes, the way Figma fluency was a few years ago. It's necessary, but it's not what makes you valuable. How you position yourself will determine your success.
Roles are blending, not just disappearing.
Designers are increasingly expected to think like PMs (defining problems, not just solutions) and like QA (catching when AI output is wrong). The narrow "I only do visual design" role has less runway than it used to. It's time to expand your skills and thinking and expand what you can contribute to the business.
Real soft skills are more valuable now
Execution is cheap now. Systems thinking, stakeholder navigation, and knowing what not to build are not and they're exactly what the dashboard example demonstrates. Anyone can prompt, but the preparation and planning that happens before AI prompts and the 'roadmap for prompting' is more important than ever.
Expand your toolkit
The last point, but probably the most important is to expand your toolkit. As I have already explained, Figma alone isn't going to cut it. Learning automation tools like n8n, web AI tools like lovable, and other services relevant to your team will come in handy to rapidly build realistic prototypes that can be tested, validated and iterated on. In fact, I would almost go as far to say say that Figma is the last tool to be used to make final touch ups, after the experience has been worked out.
What does this mean for businesses?
Earlier I gave an example of a rudimentary dashboard built entirely by AI. It's worth being blunt about what that pattern means at scale: products built entirely by AI, with no design or CX judgment applied, tend to converge on the same generic shape. Same tiles, same navigation logic, and same shallow read of what the user actually needs. That's not a coincidence, it's what happens when the systems thinking is skipped and only the output is generated.
That has a direct business consequence: if your product can be described as "we prompted our way to this," a competitor can do the same thing. The moat isn't the interface anymore. Interfaces are cheap to reproduce now. What's harder to copy is the why behind the product, the data you have access to, the workflows you understand, and the experiences layered on top of the technology.
You've got to start with the customer experience and work backwards to the technology. You can't start with the technology and try to figure out where you're going to try and sell it. I've probably made this mistake more than anybody, and I've got the scar tissue to prove it. … As we have tried to come up with a strategy and a vision for Apple, it started with 'What incredible benefits can we give to the customer? Where can we take the customer?' Not starting with 'Let's sit down with the engineers and figure out what awesome technology we have and then how are we going to market that?' And I think that's the right path to take.
That last point is the one businesses most often undervalue. Customer experience isn't a finishing touch you apply after the "real" product is built, it's been a competitive differentiator for the past 20 years, AI or not. A financial dashboard that's technically functional that nobody wants to open twice isn't a technology failure. It's a CX failure. And CX failures don't show up in a demo; they show up six months later, in adoption numbers nobody wants to explain in a board meeting.
So the strategic move isn't "use AI" or "don't use AI." It's making sure the people who understand why a product should work a certain way are still in the room and directing the tools, not being replaced by them. Companies that keep experienced designers and CX practitioners on board, and equip them with AI to move faster, will ship products that are both quick to build and hard to copy. Companies that treat AI as a replacement for that judgment will ship products that are quick to build and just as quick to lose to the next competitor who prompts better.
Where does this leave us?
AI hasn't made design or CX roles less necessary, it's made the absence of it more visible and more expensive. Previously, a poorly thought through product took months to build and months to fail publicly. Now it can take a weekend to build and weeks to fail, because everyone else has the same tools and the same shortcuts available.
That's actually good news for anyone who's spent their career learning the parts of this work that don't show up in a Figma. By knowing what question to ask before a project starts, knowing how to untangle ownership across a broken process, knowing that low usage is a symptom, not a diagnosis. AI doesn't compete with soft skills, it just makes the gap between people who have it and people who don't a lot easier to see.
Those who are coming out ahead aren't the ones who resist AI, and they won't be the ones who let it think for them either. They'll be the ones who use it to move faster on the easy stuff, so they have more time to spend on the hard stuff... the stuff that was always the actual job but they never had the time to do.