The Mirror Principle: How Human Input Shapes AI Output Through Relational Technolinguistics | Synapse