Newest papers

Can Loan Processors and Underwriters Spot AI-Edited Statements? An Empirical Study

Simiao Ren, Xingyu Shen, Tommy Duong, Ankit Raj · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.30758 · cited by 0 works (OpenAlex) · open-access PDF

German

AI-unterstützte Bearbeitung kann Darlehensdokumente produzieren, die vertraute Emittentenlayouts beibehalten, während die Identität des Antragstellers geändert wird. Wir untersuchten, ob erfahrene Unternehmer und Darlehensverarbeiter solche Fälschungen in der ersten Übersicht identifizieren. Drei Praktizierende, die über Upwork rekrutiert wurden, erhielten jeweils ein Antragspaket aus 134 gefälschten PDFs, die aus 47 mit vollem Einverständnis gelieferten echten Finanzdokumenten stammen. Die Bearbeitungen behielten Beträge, Termine und Salde; Identifizierungsunterschiede, die von der Pipeline eingeführt wurden, wurden bewusst vor der Verteilung beibehalten. Bewertern wurde nicht gesagt, dass Dateien geändert wurden und keine visuellen Ur

Machine translation (Meta M2M100); the English original follows.

Abstract

AI-assisted editing can produce loan documents that preserve familiar issuer layouts while changing the applicant's identity. We investigated whether experienced underwriters and loan processors identify such forgeries in first-pass, document-only review. Three practitioners recruited through Upwork each received one applicant package drawn from 134 forged PDFs derived from 47 genuine financial documents supplied with full consent. The edits preserved amounts, dates and balances; identifier inconsistencies introduced by the pipeline were knowingly retained before distribution. Reviewers were not told that files had been altered and returned 132 file-level judgements. Checking their explanations against the originals yielded 14 edit-supported concerns (10.6%): 12 Suspicious judgements and two Insufficient information judgements. All supported concerns involved identifier disagreements within or across documents, and no reported visual cue identified an actual edit. Thus, approximately one in ten judgements contained evidence of the forgery, despite known inconsistencies remaining in the files. These observations motivate evaluating consistency checks and independent source verification alongside human review. With three reviewer-package pairs and no genuine-file controls, the findings describe this exploratory setting rather than population-wide detection accuracy.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Underwriting, Loan, Identifier, Consistency (knowledge bases), Issuer, Actuarial science

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange