Mediation, Not Amplification

In The You On AI Field Guide

The amplification metaphor specifies three conditional expectations: that output character is determined by human input, that the human could produce the same output without the machine, and that the relationship is unidirectional. The Peirce volume tests each against the evidence from You On AI itself and finds each fails.

The character of output is not determined solely by human input. When Claude suggests the laparoscopic surgery analogy, the suggestion is not an amplified version of an idea Segal already had. It is a new idea, shaped by the machine's associative patterns, its cross-domain reach. The signal was not amplified — it was transformed.

The human cannot produce the same output without the machine. The engineer who crossed from backend to frontend development did not merely work faster. She did something she could not have done at all — accessed a domain her existing skills did not reach. The machine constituted a new capability, not an amplified one.

The relationship is not unidirectional. The machine contributes content; the human responds; each participant's contributions are modified by the other's. The signal flows in both directions, which is incompatible with the amplifier model and characteristic of genuine mediation.

Origin

The correction applies Peirce's pragmatic maxim — that a concept's meaning consists in its practical consequences — to a concept central to You On AI.

The analysis draws on Peirce's semeiotic, which provides a more precise vocabulary for describing mediation than the engineering metaphor of amplification allows.

Key Ideas

Pragmaticist test. Apply the pragmatic maxim — what practical consequences does the concept specify? Amplification's consequences don't match observations.

Mediator transforms. Every mediator introduces its own tendencies; the translator's choices, the editor's voice, the machine's statistical patterns all shape the result.

Mediator literacy. Using a mediator well requires understanding how this particular mediator transforms the signal — its biases, preferences, default patterns.

Institutional consequences. Organizations that adopt the amplification model invest in better prompts; organizations that adopt the mediation model invest in evaluative judgment.