The industry just caught up to something I've known for seven years. NNGroup's State of UX 2026 landed this week with a headline already circulating across every design Slack: AI fluency is now essential. Over half of hiring managers say they won't consider designers who can't work with AI. Figma followed with their own piece on skills for the AI era. The message is clear, learn to use the tools or get left behind.
I agree with the premise. I disagree with the framing.
What seven years of shipping AI in product actually teaches you
In 2018, I worked on the UK's first AI application in retail banking. A behavioural analytics platform that generated financial insights at scale. I spent more time reviewing what the AI got wrong than designing what it got right. That pattern repeated when I led design on LLM-powered search at super-app scale across multiple markets. And again on Claude- and MCP-driven design work in crypto. In every case, the critical skill wasn't prompting. It wasn't knowing which model to use. It was the judgment to know when the output was off, and the experience to say why.
The industry is running at AI fluency like it's a new certification. Get comfortable with the tools. Learn the prompts. Add it to your LinkedIn. Fine as far as it goes. But fluency without taste is just fast mediocrity.
Why taste beats fluency
What actually separates senior designers in the AI era isn't how much they use the technology. It's that they've seen enough bad product decisions, designed, built, shipped, and measured, to know when AI is optimising for the wrong thing. The model doesn't know that your fintech users in MENA distrust 'no fees' positioning because they've been burned before. It doesn't know that the insight it generated sounds helpful but will trigger a compliance review. It doesn't know that the onboarding flow it suggested is technically sound and experientially awful.
You know. Because you've been in the room.
The shift worth making
The shift worth making isn't about tool adoption. It's about building the kind of critical eye that makes you valuable precisely because AI can't replicate it. That means shipping more, not just prompting more. It means forming strong opinions about what good looks like, then testing them against reality. It means treating AI output the way you'd treat work from a junior designer, a useful starting point that needs your judgment on top.
AI fluency is table stakes now. The right call. But the designers who build durable careers in this era won't be the ones who used AI the most. They'll be the ones who knew when not to.
Fact Check
Every factual claim in this article, with its source.
Claim: NNGroup's State of UX 2026 declares AI fluency essential for designers.
Source: Nielsen Norman Group, State of UX 2026 report. Specific URL not captured at draft time. Verify before re-promotion.
Claim: Over half of hiring managers say they won't consider designers who can't work with AI.
Source: Nielsen Norman Group, State of UX 2026 report. Verify before re-promotion.
Claim: Figma published a supporting piece on skills for the AI era.
Source: Figma blog (2026). Specific URL not captured at draft time. Verify before re-promotion.
Unsourced statements (Jay's opinion or lived experience): Seven years of AI-in-product work since 2018; the MENA fintech 'no fees' example; the closing argument that designers who knew when not to use AI will build durable careers. These are Jay's points of view, not third-party data.