For most of my career, the pixel felt precious.
In the early days of desktop publishing, I thought of pixels as the digital cousin of ink dots on press. You earned the right to move them around. You learned color, trapping, resolution, typography. You invested in hardware that sounded like a NASA checklist. The pixel was a scarce resource, and the people who knew how to bend it to their will were scarce too.
Typesetters, color separators, prepress operators, early digital artists, UI designers, animators—we were the ones doing the careful work, turning raw tools into finished things. A good layout or interface wasn’t just “on the screen”; it was the result of time, practice, and constraints.
That scarcity is gone.
The pixel factory just went infinite
Tools like Nano Banana aren’t just “better Photoshop.” They’re something stranger: a pixel factory that runs on language.
You describe what you want and a finished screen, image, storyboard, or interface appears. Not a blank canvas. Not a half-baked mockup. A whole composition, with lighting, perspective, control layouts, iconography, and micro-details filled in at machine speed. And you can do this again and again until something “feels right.”
That has a simple, brutal implication:
The cost of manufacturing pixels is collapsing toward zero.
Need a new hero image for your site? Ten options in thirty seconds. Need a landing page layout? Prompt, tweak, ship. Need a mood board for a client? You’re drowning in versions before your coffee cools.
We’ve quietly crossed a line: pixels are no longer the product. They’re exhaust.
What happens to the pixel craftsmen?
If your job has been drawing, laying out, or animating pixels manually, this sounds terrifying—and honestly, it should at least get your attention.
I’ve watched a version of this movie before.
When desktop publishing showed up, I consulted for a typesetting house that tried to pivot into a PostScript service bureau. They had the experience, the equipment, the track record. What they didn’t have was the willingness to let go of the idea that their craft was the center of the universe. The market chose speed and control on the desktop over perfection on film, and a whole class of experts got crushed by that decision.
AI is doing something similar now, just higher up the stack.
A lot of today’s design, illustration, and UI work is painstaking pixel manufacture: pushing shapes, nudging spacing, redrawing variations. The market is about to discover it can get “good enough” versions of that in a fraction of the time and cost.
That doesn’t mean designers or artists disappear. It does mean the part of their work that’s purely about producing pixels is going to be devalued. Hard.
The work moves above the pixel
So where does the real value go?
You can think about pixel work in three rough layers:
- Manufacturing pixels Turning prompts into images, layouts, storyboards, or interfaces. This is now cheap and getting cheaper.
- Arranging pixels in a system Deciding patterns, flows, hierarchies, and rules. What’s reusable? What breaks? How do things behave over time, not just on a single screen?
- Deciding what pixels mean The story. The promise. The trade you’re making with a human being’s attention. The ethics of what those pixels are nudging people toward.
AI is eating layer one. It is rapidly learning to assist with big chunks of layer two.
Layer three is where humans still matter most.
If you’re a designer, art director, motion artist, or UI person, that’s the invitation. The job is shifting from “I make pixels” to “I’m responsible for meaning and consequences.”
- Instead of just picking colors, you’re deciding what those colors should signal to a scared patient, a confused customer, or a tired frontline worker.
- Instead of just shipping another signup flow, you’re asking what happens to someone’s data, attention, or autonomy once they step through it.
- Instead of just crafting a beautiful poster, you’re deciding what story that poster reinforces about who we are and what we value.
The tools can flood the world with pixels. Somebody still has to decide which ones are allowed to reach people and why.
From craft of the hand to craft of the lens
When pixels were expensive, craft meant mastery of the tools: knowing your software better than anyone else, understanding every little trick in the menu.
In a world of cheap pixels, craft looks more like this:
- Pointing the lens. Knowing where to aim the tools and where not to.
- Setting constraints. Deciding what the system is not allowed to generate, even if it can.
- Curating at speed. Looking at 50 AI-generated options and choosing the one that actually respects the story, the user, and the context.
- Building substrates, not just surfaces. Helping shape the underlying systems, data, and rules that the UI or image is sitting on top of.
The irony is that the people who are best positioned to do this are the very folks who feel most threatened by “pixel automation.” They know what bad design looks like. They know what lazy composition feels like. They know when something is visually correct but emotionally wrong.
They just have to be willing to move up a rung.
Why I’m writing this down now
I’m writing this partly as a note to myself.
I started out in a world where knowing how to push pixels was a superpower. I watched whole shops disappear because they treated that skill as sacred instead of temporary.
We’re at another inflection point. Pixel manufacturing is becoming disposable. The value is sliding toward direction, judgment, and responsibility.
In future pieces, I want to explore this further:
- What does a healthy career path look like in a world of disposable pixels?
- How do we teach younger creatives to focus on meaning, not just output?
- What does ethical design look like when any surface can be generated on demand?
For now, I just want to put a stake in the ground:
Pixels are no longer the scarce thing. Your taste, your story, and your standards are.