Opus 5's disproof of Conjecture 349 — a result listed as proved in the literature — has implications beyond the specific conjecture. The Jiang 2012 proof in Utilitas Mathematica was accepted by the mathematical community as valid; Opus 5's infinite caterpillar family demonstrating systematic violation suggests either an error in the 2012 proof or a misinterpretation of the conjecture statement in the DeLaViña-Larson-Pepper-Waller survey. Opus 5 explicitly noted "the statement appears verbatim in the survey, so the parse isn't in doubt" — ruling out the simplest alternative explanation. The result demonstrates that automated mathematical reasoning systems can not only tackle open problems but also challenge accepted results, functioning as a verification layer for the mathematical literature itself. This represents a qualitatively different capability than solving open conjectures: it requires sufficient confidence to contradict published, peer-reviewed work.