Review-attributed evidence: fabricated model-generated references clustered in disciplinary journals in a student verification exercise¶

Claim · All claims
Evidence · 1 study · theoretical r3 · q2
Currency · newest study 2026 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? In a classroom verification exercise reported by Sevryugina and Vargas, 51.3% of 456 model-generated references checked by students proved real and correctly cited, with fabricated ones clustered in disciplinary journals. → Muminova 2026

Evidence¶

Muminova 2026¶

Muminova, Z. O., Mamatkulov, D. M., Turokulova, R. I., Rakhmonova, N. T., Shavqiyev, O. S. O., & Zoirova, S. U. (2026). The nature of science in the age of generative artificial intelligence: Epistemic authority, materiality, and a synergetic response. EURASIA Journal of Mathematics, Science and Technology Education, 22(10), em2943. https://doi.org/10.29333/ejmste/19481

q2 · i? · theoretical · r3

The review reports this classroom verification exercise, citing Sevryugina and Vargas (2026); the study's own design is not described beyond the quoted sentence. It is used as illustrative evidence that fabricated content is not an occasional accident and that assertiveness is weakly coupled to accuracy.

"In a classroom verification exercise in which students checked 456 model-generated references against the literature, 51.3% of those references proved to be real and correctly cited, and the fabricated ones clustered in disciplinary journals rather than in high-profile multidisciplinary ones"

Discussion¶

Related Claims¶

• AI output errors are frequent: 44.5% fabricated references in drafted research proposals and 5–13% unsafe chatbot medical answers — related
• Conservative manual verification identified 30 hallucinated references across 14 papers in 2025 and 2026 computing education venues — related
• Of 30 verified hallucinated references, 13 appeared entirely fabricated and 17 were hybrid references combining a real title with fabricated authorship or metadata — related
• Poorly grounded GenAI self-study risks fabricated content and inferior learning relative to curated resources — related
• AI-generated ethics exercise materials carried external citations that did not survive checking, ranging from misattributed to apparently fabricated — related
• AI-assisted scientific writing assignments risk fabricated references and altered chemical meaning — related
• ChatGPT fabricates references: the authors caught it inventing a non-existent 'Ribeiro and Vala, 2020' article with full bibliographic details — related
• Evidence types differ systematically in probative value: detection outputs cluster at weak inferential force while admissions, observed exam behaviour and fabricated references carry the strongest profiles — related
• Experimental-group students detected AI errors and enacted verification and ethical self-regulation as routine practices — related