Field Notes
If Creation Is Free, What’s Worth Creating?
As AI pushes creation costs toward zero, attention, taste, perspective, and meaning become the scarce layer.
Lately I’ve been thinking about what becomes valuable when creation itself is no longer scarce.
AI is pushing the cost of creating things toward zero. Not just text, but music, images, video, software, brands, products, and eventually things we haven’t even thought of yet. We’re moving toward a world where almost anything can be generated on demand and at a quality level that would have seemed impressive only a few years ago.
If that’s true, then the interesting question isn’t whether AI can create. It’s what becomes scarce once everyone can.
I keep coming back to companies like Disney and Coca-Cola.
Both are incredibly successful and highly optimized for what they’re trying to do. But it would feel strange to argue that Coke is the best possible drink or that every Disney film represents the peak of storytelling.
That’s not the point.
Coke is designed to be broadly appealing and consistently enjoyable to a huge number of people. Disney often optimizes for stories that can resonate across cultures, generations, and audiences. They’re not necessarily maximizing for depth or originality. They’re maximizing for scale.
I think a lot of people frame the AI conversation incorrectly because they assume there’s a single market for creative work. Usually the discussion gets reduced to something like human creativity versus AI creativity or art versus mass-produced content.
But that’s not how people actually consume things.
Most people move between both worlds all the time. You might spend one evening listening to an album that feels deeply personal and meaningful, then spend the next day listening to familiar background music while working. You might watch an independent film one week and a Marvel movie the next. Those experiences serve different purposes.
What seems more likely to me is that AI accelerates a split that was already happening.
On one side is passive consumption. Personalized feeds, generated music, endless entertainment, and content optimized for whatever mood you’re in at a particular moment. AI will be extremely good at this because it’s fundamentally an optimization problem. Most people will engage with this layer constantly because convenience is hard to compete with.
On the other side is intentional consumption. Things people seek out because they care about the creator, the perspective, the story, or the ideas behind it. Not necessarily because the work is technically better, but because it feels like there is a real person expressing something specific.
My guess is that this second category becomes more valuable as everything else becomes abundant.
A lot of the anxiety around AI seems to come from this shift. Not just for artists, but for anyone who creates things.
For a long time, the act of creating something was a signal in itself. Recording an album, producing a film, writing a book, or building a product required enough effort and coordination that the existence of the thing implied a certain level of commitment and intent.
That signal is weakening. We’re heading toward a world where more things get created, more quickly, and at a higher average quality than ever before. That’s exciting, but it also means the amount of noise increases dramatically.
As a result, I don’t think technical execution remains the scarce resource.
Perspective, taste, judgement, and the ability to decide what is worth creating in the first place becomes scarce.
Music is probably one of the clearest examples. Long before generative AI arrived, software, plugins, sample libraries, presets, and distribution platforms dramatically lowered the barriers to production. AI is simply continuing that trend.
Today it’s easier than ever to make something that sounds professionally produced. However, it’s still surprisingly difficult to make something that genuinely resonates with people. I suspect that difference becomes more obvious, not less, as generated content becomes effectively infinite.
None of this means mass-produced content disappears. If anything, it probably becomes even more dominant. Most people will continue choosing things that are familiar, reliable, and immediately satisfying most of the time.
But when everything is easy to create, the things that stand out are less likely to be the ones that required the most effort and more likely to be the ones that reflect a distinctive point of view.
I don’t think AI replaces human creativity. I think it removes creation as the primary bottleneck.
Once that happens, attention, taste, judgment, and meaning become much more important than the mechanics of making something.