How to build an AI moat.
Moonlabs is the operator-led AI Academy in Derby. We run three live companies — Homemove, home.co.uk and homedata.co.uk — and we teach twelve students per cohort to ship a real AI product, sell it to a real customer, and raise on it. Three pillars: Coding, Commercials, Investment. Twelve weeks. £6,000.
Moonlabs is what we are. Two operators — James Freestone and Louis O’Connell-Bristow — who run Homemove, home.co.uk and homedata.co.uk. We have answered the “what stops OpenAI doing this?” question across our own businesses dozens of times — in front of investors, in front of customers, at 11pm in front of each other. The moat-construction framework on this page is the one we have used to win those conversations.
The Academy is what we do. A twelve-week, in-person, twelve-student cohort in Derby. You build a real AI product. You sign a paid pilot on it. You write a deck and a financial model. You leave with a deployed system, a paying customer reference and a live investor pipeline. Coding, Commercials, Investment — the three pillars taught in equal weight every week. Defensibility runs through all three; this page is the explicit framing of that work.
Why this page exists. Every AI founder in 2026 has heard the question. The honest answer is: nothing stops OpenAI eventually, for any sufficiently general capability — the moat must come from somewhere else. AI moats in 2026 come from four places: distribution, data, workflow, brand and trust. Almost every defensible AI company has at least two. You leave the Academy with a defensibility story that holds up — or you have killed the idea early enough to save six months.
Coding · the data and workflow moats, built
Data the foundation models cannot reach — ingested, cleaned, structured, with feedback loops that improve your product specifically. Deep workflow integration — agents wired into the buyer’s systems via MCP, retrieval over their archives, eval suites that prove the integration works. A deployed product with at least one structural moat baked in, by week twelve.
Commercials · the distribution and trust moats, earned
Distribution beats capability. Most AI features in 2026 are won by the company already in the buyer’s workflow, not by the smartest model. The paid pilot signed in week six is the start of the distribution moat — reference customer, case study, reputation in the buyer’s peer network. A second moat (distribution or trust) by week six — the “why you” answer that survives a cynical investor question.
Investment · raising on defensibility, not capability
Investors back companies with at least two moats. Palantir (workflow + trust), Glean ($4.6bn, distribution + data), Sierra ($4.5bn, workflow + trust), Cursor ($2.5bn, distribution + brand), Hebbia ($130m, workflow + data), Harvey ($1.5bn, trust + data) — every loud AI fundraise of 2025-26 has at least two of the four moats visible in the first three slides. A live investor pipeline by demo day, with the moat story pre-rehearsed.
Common questions.
Is "thin wrapper" a real risk to my startup?
Honest answer: if your product is a thin wrapper, yes. If your product is the wrapper plus distribution, plus data, plus workflow, plus brand — almost no. The conversation matters more than the diagnosis.
What if my moat is just "we move faster"?
Speed is a moat for ~18 months, never longer. We will not let you bank on it as your only one — the Academy is built around finding the durable second moat alongside speed.
How do I answer the OpenAI question in an investor meeting?
Name the threat honestly. Name what protects you specifically. Name what would have to be true for OpenAI to win. The pitching page covers this in detail.
Can I build my way to a moat retrospectively?
Sometimes. The companies that find a moat after starting tend to do so via the workflow or data path, rarely the brand path. We will help you reverse-engineer one if your starting wedge does not have one yet.
What if I just want to build a small profitable AI product without a moat?
Genuinely valid — not every AI company needs to raise. The solopreneur framing covers that path. The moat question is most relevant if you intend to raise or scale a team.
More Academy entry points.
The Academy is one course with many doors. Each of these pages is a different entry point into the same twelve weeks.
How To Hire Your First AI Engineer
How To Raise For An AI Startup
How To Become An AI Engineer
How To Pitch An AI Startup
How To Validate An AI Startup Idea
How To Build An AI Startup
Build it. Sell it. Raise on it. In twelve weeks.
Tell us the wedge you would defend and which two of the four moats you would aim for. James and Louis read every application personally and reply inside the week.
© 2026 Moonlabs Incubator. All rights reserved.