n.V6-4.07 | THE LINE WE DO NOT CROSS
01| Every AI capability we consider gets one question before adoption, and it is not "does it work" or "what does it save". It is: does this serve self-perpetuation or replace it?
02| If AI accelerates training without removing the human teacher, it serves. If it replaces the teacher, the loop has nothing left to replicate. If it surfaces patterns that help a Maveriq decide better, it serves. If it decides for her, it has removed the thing the platform exists to build. If it scans thousands of projects for compliance problems, it serves. If it overrides a community's judgment about what matters on their own street, it has broken the only commitment that makes us different from what came before.
03| The test cuts both ways, which is why it is worth having. It licenses us to use AI aggressively — more aggressively than most of our sector will be comfortable with — everywhere the model permits. And it forbids AI exactly where cost pressure will most tempt us.
04| Four things stay human. Agency: residents lead their own transformation, which means they choose the projects, and an AI that chooses for them has replaced the point. Trust: built by presence, consistency and shared experience over years; a machine cannot accumulate it and a Maveriq who has lived there for five years already has. Judgment in conflict: when two people in a team fall out, when a project is failing, when a community goes quiet on us, the answer requires reading what is not being said. The two-generation play: children grow up watching capable adults, and the inheritance we are betting on is a neighbour who ran a budget, a parent who designed something that got built. An agent in that role produces no inheritance at all.
05| These are not statements about what AI is incapable of. They are choices. If the technology advances to the point where it could plausibly do all four, the choice stands. A platform that scales by removing the people it exists to serve has not scaled; it has changed subject.
06| What we sacrifice, stated plainly: money and speed. Keeping teaching, conflict and decision-making human is the most expensive commitment in this volume and the one that most limits how fast we can grow. We are choosing not to be the cheapest possible version of ourselves. Anyone reading our cost projections should read them with this line in them, because it is the reason the numbers are not lower.
07| And the funder should understand what they are backing. Not the technology. The people using it.
