You Have Good Taste. Can You Explain It?
The shift from static thinking to deterministic design is where the next era of product UX begins.
Something I keep noticing in the field: designers who are genuinely talented, experienced, and thoughtful. Still thinking about their work in static terms.
A screen. A component. A template. A handoff.
It’s not that these designers are bad at their jobs. Most of them are excellent. The problem is they’re solving for a world that’s already shifting underneath them. They’re making beautiful, considered decisions every day and storing almost none of that knowledge anywhere an AI could use it.
That’s the gap I want to talk about.
There are five stages of design maturity. Most designers are stuck in the first three.
Here’s a framework I’ve been developing to map where the industry actually is versus where it’s going.
Stage 1: Interface Design Designers create individual screens. Layout, typography, hierarchy, color. This is where most of us started.
Stage 2: Component Design This is where design systems and atomic design emerged. Reusable elements, UI kits, libraries. A real step forward in thinking systematically.
Stage 3: System Design Designers define tokens, rules, and composition patterns. Layout systems, component relationships, composition patterns. This is the current ceiling for most serious practitioners.
Most designers are still operating in stages 1-3. AI is rapidly advancing in these stages. That’s not an opinion. It’s where the work is going.
Stage 4: Adaptive Design Experiences adapt to content, data, device, and user intent. Designers aren’t composing pages. They’re defining content ranges, layout rules, and behavior constraints. The output isn’t a static screen. It’s a set of conditions the interface responds to.
Stage 5: Generative Design AI introduces something fundamentally new here. Designers define the rules that generate experiences. Instead of designing pages, you’re designing:
Experience rules
System constraints
Intent models
Composition logic
Deterministic design
That last one is worth defining. Deterministic design means designing rules and expected outputs in advance, so AI can execute predictably.
You’re not hoping AI makes good decisions. You’re building the system that makes the decisions for it.
This is where modern product UX is heading. And it requires a different kind of designer.
The real skill gap isn’t tools. It’s articulation.
Most designers I talk to have excellent taste. Strong instincts. Clear opinions about what makes something work. Years of pattern recognition built up through doing the work.
But ask them to explain why something is right, in terms precise enough for an AI agent to act on it, and it falls apart. They’ll say things like “it just feels off” or “this needs more breathing room” or “the hierarchy isn’t landing.” Valid observations. Completely unusable as instructions.
This is the gap that matters right now.
It’s not enough to be a good designer anymore. The skill gap is whether you can articulate what you know well enough for an agent to act on it.
Your taste, your standards, your judgment. The designers who can translate what they “just know” into something an AI can execute will define the next era of product design.
Taste without articulation used to be fine. You’d sit in a design review, point at the thing, and say fix it. A human would interpret your feedback and make the change.
AI can’t do that. It needs a different kind of input:
Rules
Constraints
Ranges
Intent
It needs you to have already done the thinking before it starts working.
What the mindset shift actually looks like
This isn’t abstract. Here’s the practical version:
Instead of Executors → Tastemakers and shape shifters
Instead of Reacting to requests → Orchestrating experiences
Instead of Pixel perfection → Intent-driven flexibility
Instead of Shipping components → Defining intelligent composition rules
Instead of Building static pages → Designing adaptive systems
Instead of Exact content → Acceptable ranges and guardrails
Instead of Templates → Intelligent frameworks
Notice what connects everything on the right side: explicitness. You can’t orchestrate an experience without knowing what it’s supposed to do. You can’t define composition rules without being able to state them. You can’t set guardrails without knowing where the edges are.
This is the work. Making the implicit explicit.
The fog most designers are in
I don’t think most designers are ignoring this shift. I think they’re genuinely uncertain about where things are going and what to do next.
Here’s the thing: everything you’ve built up, your taste, your instincts, your years of knowing what “good” looks like, that’s the asset. It doesn’t go away. It becomes more valuable. But it needs to be externalized.
The ones who get there won’t be the most technical. They’ll be the most articulate.
You don’t just need to know what good looks like. You need to know why, at a level precise enough to write it down. That’s a new ask. But it’s one that plays directly to what great designers have always been good at: making decisions, holding a vision, knowing where the edges are.
Stages 4 and 5 aren’t a threat to that. They’re where it finally gets put to use.
The question worth sitting with
It’s not “will AI replace designers?”
It’s “can you explain your taste well enough that it survives you?”
If the answer is yes, you’re not a candidate for replacement. You’re the person who trains what comes next.
I write about design systems, AI, and where the industry is actually heading. If this landed, follow me on LinkedIn. What’s your take? Drop it in the comments.