I came across a reel recently that genuinely excited me. A founder (@askcatgpt) walking through a soon-to-be 2,200 square foot creative studio built around a single premise: make products her audience actually asked for. One experiment a month. Some will become a real business, some will fail publicly. All of it shown — the prompts, the tools, the reasoning, the process. The audience decides what gets built.
It's not a revolutionary concept. But the genuine transfer of creative authority to the people who will actually use what gets made — that's pretty cool.
When anyone can build anything
AI has made it easier than ever to turn an idea into a product. The barrier between concept and reality — which used to require significant capital, a development team, and months of lead time — has compressed dramatically.
When anyone can build anything, a lot of people will. And a significant proportion of what gets built will provide little to no real value — not because people are careless, but because moving fast doesn't tell you what someone actually needs. There is a version of the near future where the market fills with products that are technically functional, competently designed by AI standards, and almost entirely disposable.
In that environment, users become more discerning, not less. When everything is available and most of it is fine, people reserve their attention — and their money — for the things that actually improve their day. The lower the cost of production, the higher the bar for real value.
Which is why direction matters so much right now. Not just the direction of the design, but the direction of the whole endeavour — knowing whose problem you are solving, why it matters to them, and whether the thing you are building is really worth their time.
What community-led building actually looks like
What makes Cat's model so interesting is that it answers the direction question before anything gets built. The community is not consulted after the fact — they are the brief. Real people who have already demonstrated trust in the person building for them, signalling what they want to see exist in the world.
That is about as strong a starting point as product design gets. It does not eliminate risk — some experiments will still fail, and she is upfront about that. But it removes the single biggest failure mode in product development: building something nobody asked for.
The other thing worth noting is the transparency of the process. Not polished updates or milestone announcements, but the actual work — the AI prompts, the tools, the prioritisation decisions, the dead ends. At a moment when so much product development is hidden behind finished surfaces, seeing how something gets made feels more valuable and more rare than it has in a long time. A process willing to be seen feels rare right now.
Where AI fits in
This model only works at the speed it does because of AI. Monthly experiments across hardware, software, and digital products — built transparently, shipped publicly, iterated on in real time — would not have been feasible for a small team a few years ago without significant capital behind it. AI collapses the distance between idea and testable reality in a way that changes what is possible.
But the AI is not the creative. It is the velocity engine. The insight, the taste, the community relationship, the judgement about what is worth making — those are entirely human. Remove the human direction and you do not get interesting monthly experiments. You get fast output that nobody asked for, in a market that is already full of it.
What AI-assisted work at its best actually looks like: the community signals, the founder interprets, the tools make it buildable, and the loop closes fast enough to matter. That's the version worth paying attention to.
What this means for how we think about product design
The most important takeaway from this model is not the format — not the monthly experiments or the public failures. It's the underlying commitment to putting real people at the centre of the process, and getting seriously creative about how to understand, explore, and solve their problems.
That has always been what good product design looks like. What is different now is the stakes. In a world where building is cheap and fast and increasingly automated, the human work — the listening, the interpreting, the caring about whether something actually improves someone's day — is what separates the products worth making from the ones that simply exist.
When anyone can build anything, the question of what is worth building becomes more important, not less. Direction, real user needs, and genuine creative curiosity are not soft skills or nice-to-haves. They are the difference between a product that earns a place in someone's life and one that disappears into the noise.
