Shared reality

LLM notes receive more positive ratings than human notes across raters with different political viewpoints, suggesting the potential for LLM-written notes to achieve the cross-partisan consensus.

A real deployment study of LLM-written Community Notes finding they outperform human notes across partisan lines — evidence AI can strengthen, not just erode, crowd-based fact-certifying procedures.

AI notes achieved a VSR of 88.8 percent compared to 68.5 percent for human-written notes (chi-squared P < 0.001).

Independent seven-month evaluation of X's AI Note Writers pilot showing AI notes reach 'helpful' status at double the human rate — converging with the MIT field study on AI augmenting community fact-checking at scale.

When these systems are robust, they support collective decision-making under pressure. When they fail, crises do not merely become harder to manage; they accelerate and spiral.

An author of the canonical Turing epistemic-security report updates the infrastructure frame for the AI era, naming AI-enabled manipulation among seven priority interventions.

habits and turns of phrase that are rare on Wikipedia but common on the internet at large

Documents Wikipedia's 'Project AI Cleanup' and its Signs of AI Writing guide — an epistemic institution building working procedures to defend itself against AI slop.