Editing Openai/6973e373-2b8c-800d-84b7-d7001cfc6bac (section)

===== Axiom 9 — Good faith is not a feeling; it’s a pattern of repair behaviors. =====
Good faith shows up as quoting, specifying, answering, conceding, narrowing claims, stating update-conditions.

Axiom 10 — Bad faith can be functional without being conscious.
People can sincerely believe what they’re saying and still behave in ways that function as derailment.

Axiom 11 — Substitution moves are the signature.
When asked to engage premises, the person substitutes: tone labels, meta labels, credential leverage, narrative reframes, ridicule, or exit moves.

Axiom 12 — The “path of least resistance” selects the cheapest move that protects the self-story.
The mind doesn’t need to be evil; it only needs to be avoidant of the self-correcting cost.

Axiom 13 — “Time/attention limits” are only exculpatory before the repair test.
If someone keeps replying but won’t do minimal repairs, it’s no longer about time; it’s about payoff.

Axiom 14 — Overgeneralization is a reliability leak.
Universal claims require universal-level justification; otherwise they are epistemically negligent.

Axiom 15 — Premise discipline beats conclusion performance.
A valid conclusion with missing premises is still an unreliable mind at work.