The most valuable sentence a technical partner can say is often "you do not need AI for this." It rarely gets said, because saying it means turning down work. We think it is exactly the sentence worth paying for, so it is worth explaining when it applies.
The demo is not the problem
A convincing demo proves that a model can produce a plausible output once, under conditions you chose. It says almost nothing about whether the same approach will hold up against real inputs, real edge cases, and a business that needs the answer to be right most of the time. Confusing the two is the single most expensive mistake we see.
Three signs you are reaching for the wrong tool
AI is usually the wrong answer when one of these is true:
- The problem is deterministic. If there is a correct answer that a rule, a query, or a formula can produce, a probabilistic model only adds cost and uncertainty.
- You cannot tolerate wrong answers, and cannot check them. If a mistake is expensive and you have no cheap way to verify the output, you are buying risk, not leverage.
- You do not have the data or the feedback loop. A model with nothing to learn from, and no signal telling you when it drifts, is a liability dressed as an asset.
What to do instead
Sometimes the honest recommendation is a boring one: a well-placed rule, a cleaner data pipeline, a better interface, or simply fixing the process that generates the problem. These are less exciting to announce and far more reliable to run. When AI genuinely is the right tool — and often it is — the case for it survives this kind of scrutiny easily.
Good engineering starts by choosing the simplest thing that solves the problem. That discipline is the whole point of a discovery workshop: to find where AI pays, and to say so plainly where it does not.