Skip to content
arghiv

> Rhetoric > Oracle-Mediated Rhetoric > arghXiv:2601.05529

Detection of Machine-Generated Text by Participants in Adversarial Exchanges

R. Adeyemi

Abstract: Participants were asked to identify which of an opponent's messages had been produced by an automated system. Accuracy was 49.6 per cent against a chance rate of 50 per cent; mean reported confidence was 91 per cent. Accuracy did not improve when participants were told that half the messages were machine-generated, nor when they were told, truthfully, that none were.

Comments:
13 pages, 2 figures.
Subjects:
Oracle-Mediated Rhetoric (rhet.AI) (primary class); Eristics (argu.ER); Verification and Provenance (epist.VER)
Cite as:
arghXiv:2601.05529v1 (BibTeX)
Submission history:
From: R. Adeyemi [view email]
[v1] Tue, 13 Jan 2026