AI-built business
Revenue comes from the AI capability: subscriptions, licences, usage, integrations, training or implementation. It competes in a fast-moving technology market.
End Up Here · Working question
Making money with AI
Where does the
money actually
come from?
AI can be the thing you sell. Or it can be the thing that lets you sell something else. Those are related, but they are not the same business.
The distinctionIs AI the product, or is AI the leverage that makes a useful proposition possible?
Two approaches
AI as product.
AI as leverage.
AI as product means making money directly from AI-related value: a model, an agent, a software tool, an API, a workflow product, training, implementation or access to compute. The customer is paying because the AI capability itself is what they want.
AI as leverage starts with a more ordinary proposition: a service, a shop, a publication, a local solution, a craft, a piece of research or a business that solves a recognisable problem. AI helps one person or a small team research, design, make, market, deliver and improve it.
The useful question: if the AI disappeared tomorrow, would there still be a customer problem worth solving? If yes, the business may be AI-enabled rather than AI-dependent.
Compare the foundations
AI-enabled business
Revenue comes from a customer’s outcome: a repaired object, a better meal, a useful report, a cared-for garden, a delivered service or a product people want.
Hybrid
Some businesses sell a tool and use it themselves. The durable advantage may be the combination of a real domain, trusted relationships and better use of AI.
A simple test
Follow the value
back to the customer.
Start with the need. Who has a problem, desire or constraint? What are they already trying to do? What would a noticeably better result be worth to them?
Name the exchange. What exactly is being bought: access to intelligence, or an outcome produced with intelligence? Can the offer be explained without the word “AI” doing all the work?
Find the human edge. Trust, taste, judgement, local knowledge, responsibility and persistence do not vanish because the production process becomes faster.
Price the result. Faster production can improve margins, lower prices, increase quality or make a previously impossible small business viable. It does not automatically create demand.
The opportunity
Make more things
possible for one person.
The exciting prospect is not that everybody must become an AI entrepreneur. It is that a person with a good idea, a modest budget and a particular knowledge of the world may be able to put that idea into practice.
AI can act as researcher, organiser, designer, assistant, translator, analyst, prototype-maker and patient first employee. The proposition still needs to be useful, trusted and wanted. But the threshold for trying it can fall.
That could mean a tiny local service, a specialist catalogue, a new kind of publication, a repair network, an educational offer or an existing business made more responsive. The income is still grounded in the thing people value.
Questions to keep asking
What is the customer really paying for? What would remain if the AI tool changed? Does automation make the work more humane, or merely push effort and risk somewhere less visible?
Who owns the useful system? Who is accountable when it is wrong? And does the new capability widen the number of people who can act, or simply concentrate another market around a few platforms?