The Part of Design AI Can’t Do: Why Your Judgment Still Matters
Reading time: 8 minutes and 52 seconds
You have an idea for some quilt blocks that you love. Or you have a pile of blocks from a “block of the month club” but you didn’t care for the setting they suggested. Or you just have a pile of orphan blocks from things you started but decided weren’t for you, or blocks from workshops that never went anywhere. What are you going to do with them? It feels wasteful to leave them sitting in a drawer.
So you start thinking about settings. Layouts. Sashing. Borders. Negative space. You pull fabric. You move things around on the design wall. You take a photo with your phone. You squint at it. Something still isn’t quite right, but you can’t put your finger on what.
That’s a very normal place to get stuck. And it’s also the moment when a lot of people start wondering whether AI might help—or whether it’s going to steamroll their creativity.
Let’s clear something up right away: AI is not here to “do your art for you.” It doesn’t have taste. It doesn’t know what’s meaningful. And it definitely doesn’t know what you like.
What it can do is something much more boring and much more useful: it can help you look at more options, faster, so you can make better decisions.
Creative Work vs. Grunt Work
When you’re making a quilt (or any creative thing, really), there are two very different kinds of work happening.
Creative work is:
- Deciding what you want the piece to feel like
- Choosing what belongs and what doesn’t
- Knowing when something feels balanced, tense, calm, or finished
- Saying “yes” to this and “no” to that
This is taste. This is judgment. This is the part that comes from experience and from paying attention.
Grunt work is:
- Trying ten variations of the same layout
- Testing different border widths
- Seeing what happens if you change the scale
- Swapping color families just to check
- Exploring ideas you already suspect won’t work, but want to rule out
None of that is glamorous. It’s necessary, but it’s repetitive and time-consuming.
AI is very good at the second category.
It is not good at the first one.
What AI Actually Does in a Creative Workflow
Stripped of hype, here’s what’s really going on:
- It generates options based on parameters you set
- It speeds up iteration so you can see more possibilities faster
- It cannot make aesthetic judgments
- It cannot tell you what’s good
- It still requires you to know what you’re aiming for
Think of it like a very fast, very literal assistant who will happily mock up a bunch of versions of your idea—but has absolutely no taste.
It doesn’t know which one is better.
It doesn’t know which one fits your style.
It doesn’t know your audience, your values, or your goals.
If your direction is vague, the results will be vague.
If your direction is thoughtful, the results get more useful.
AI doesn’t replace your judgment. It makes your judgment more important, because now you’re choosing between more options, not fewer.
Where This Is Actually Useful
Let’s keep this grounded in real quilting problems, not tech demos.
1. When You’re Stuck on a Specific Design Problem
You like your blocks. The center works. But the setting feels off. The borders are either too heavy or too empty. The whole thing feels… unresolved.
This isn’t a lack of creativity. It’s a lack of visibility. You can’t easily see alternatives without committing time, fabric, and wall space.
This is where generating options can help—not to find “the right answer,” but to spark a direction you can develop further.
Your role stays the same:
- “That’s too busy, but I like the negative space idea.”
- “The scale is wrong, but the asymmetry is interesting.”
- “I hate that, but now I know what I don’t want.”
That reaction? That’s the real creative work.
2. When You Have Too Many Ideas
Some quilters don’t get stuck because they have no ideas. They get stuck because they have too many.
You can imagine three layouts, five colorways, and at least two different moods for the same quilt. And now you’re frozen, because choosing one means letting the others go.
Being able to quickly see those options makes deciding easier, not harder.
It turns “I think this might work” into “I’ve seen the alternatives, and this one actually says what I want it to say.”
Again, the tool doesn’t decide.
It just puts the options on the table faster.
3. When You’re Burning Time on Repetitive Design Tasks
If you design patterns, teach, or make work for sale, you already know how much time gets eaten by setup, variation, and testing.
Speed here isn’t about cutting corners. It’s about protecting your energy for the parts that can’t be automated: refinement, messaging, and final decisions.
The difference between designing for two hours and designing for twenty minutes isn’t laziness. It’s leverage.
A Real Example: The Swastik Quilt (It's Not What You Think, I Promise)
So What Is AI Actually For?
I made a piece for my friend Akshay in India for his family: a quilted swastik.
Context matters here. In India and many other cultures, the swastik (or swastika) is an ancient symbol of good fortune, well-being, and auspiciousness—long before it was appropriated and corrupted by the Nazis. If you want a clear cultural explanation, try Britannica for a good one.
The central design worked. I loved it on its own. But I couldn’t figure out how to finish the piece.
Originally, I thought I’d add traditional quilted stars around it. When I tried that, it felt wrong. Too busy. Too much. The center didn’t need competition.
So I took a photo of the flimsy, uploaded it, and asked for ideas on how it could be finished.
Not because I wanted a machine to design my quilt.
But because I wanted to see options I could react to.
Some were bad. Some were boring. A couple made me go, “Oh… that’s interesting.” One of those nudged me toward the solution I actually used.
The tool didn’t choose.
It didn’t understand the symbol.
It didn’t know the emotional weight of the piece.
I did.
It just helped me get unstuck faster.
And here it is quilted:
The Part AI Can’t Do
AI cannot:
- Know your audience
- Understand your values
- Recognize when something feels emotionally right
- Decide what’s tasteful, meaningful, or appropriate
- Take responsibility for the final result
It also can’t tell you when to stop.
That moment when you say, “Yes. This is it.”—that’s human judgment. Always has been. Always will be.
If anything, having more options on the table makes curation more important, not less.
Choosing becomes the craft.
So What Is AI Actually For?
- Replacing your taste
- Replacing your voice
- Making the “right” choice for you
- Turning creativity into a slot machine
AI Is for:
- Unsticking creative blocks
- Exploring variations quickly
- Testing scale, balance, and direction
- Reducing friction between an idea and seeing it
- Saving your time for decisions only you can make
Specifically:
- Know what problem you’re actually trying to solve
- Know what to ask for and why
- Know how to evaluate what comes back.
- Know what to keep, throw out, or refine
That’s the work I teach in workshops—not “push this button and art falls out,” but how to use these tools inside a real creative process without giving up authorship or control.
Because you’re still the designer.
Always.
The tool just helps you look at more sketches.
More Like This
Outside the United States?
Check Amazon for distribution availability in your country.
Hey there, I’m Theresa, The AI Quilter. I’m here to help you connect technology and creativity in ways that feel human, creative, and completely your own.
Explore More Topics
