Every durable mid-market business is built on encoded judgment: the estimator who prices complex jobs accurately, the underwriting instincts, the service knowledge that lives in twelve senior heads. For decades that expertise was the moat, protected by the years it takes to grow. AI inverted the logic. Judgment that repeats across many similar decisions is precisely what modern AI encodes well, which means the deeper and more repeatable your expertise, the more valuable a target it makes. Your moat and your threat are now the same fact.
What the challenger sees when it looks at your category
An AI-native challenger does not attack your whole business. It picks the expertise-dense, repetitive slice, quoting, claims triage, standard advisory work, diagnostic routing, encodes a good-enough version of the judgment, wraps it in faster response times, and prices it against your cost structure. Good-enough plus instant plus cheaper wins a share of every category it enters; taxi medallions and travel agencies learned the pattern from earlier waves of disintermediation. The challenger's version of your judgment does not need to beat your best senior person. It needs to beat the wait for your best senior person.
The uncomfortable audit
Three questions locate your exposure. Which of our revenue streams is expertise applied to broadly similar cases, over and over? For that stream, how much of the price defends the judgment itself, and how much defends the delay and overhead around it? And if a competitor offered the same judgment at 70% quality, instant, and 40% cheaper, which of our customers would try it? Executives run this audit and go quiet at the third question, because the honest answer is: the customers we make wait.
The move: encode your own expertise first
The answer to moat inversion is not defending the old moat harder. It is running the encoding play on yourself before a challenger runs it on you, with an advantage no challenger has: your experts, your case history, your customer trust. Capture the senior judgment into workflows where AI applies it to the routine cases instantly and your experts hold the exceptions, the relationships, and the frontier. Your cost structure drops toward the challenger's while your quality stays above it, and the expertise stops retiring with the people who carry it. The apprenticeship effect compounds: juniors ramp against the captured judgment instead of waiting years for exposure.
Timing decides who runs this play on whom, which is why the survival math matters: the encoding is happening in your category at some rate, and your counter-rate is a choice. The full treatment, with the Power Curve that puts a number on it, is in The 2026 AI Strategy, free, and the frameworks behind it are published open. The expertise took decades to build. Deciding who monetizes it next should not take more than a quarter.
