Fred Brooks's 1975 empirical observation that adding people to a late software project makes it later — because communication overhead scales as n(n-1)/2 while productive capacity grows linearly.
Brooks's Law is the most famous single claim in software engineering, and the one whose apparent violation by AI-augmented solo building has generated the most commentary. The law does not say that teams are bad; it says that communication overhead is real, that it scales nonlinearly, and that late projects already operating near capacity cannot absorb additional communication burden. A team of ten requires 45 communication channels; a team of twenty requires 190. Doubling the team quadruples the overhead. Brooks observed this pattern across the IBM System/360 project and codified it as a principle that has held for fifty years. The AI transition does not refute the law — it confirms it in the strongest possible form by showing that the optimal team size, for a significant class of work, is one.
Brooks's Law
In The You On AI Field Guide
The mechanism Brooks identified is not a property of software but of human cognition and organization. Any task that requires coordination among multiple agents incurs a coordination cost,