Newest papers

Humans Correct Their Judgments Through Debate, but Weaker AI Models Do Not

María Victoria Carro, Denise Alejandra Mester, Facundo Nieto, Oscar Stanchi, Guido Ernesto Bergman, Giovanni Franco Gabriel Marrafini, Trinidad Borrell, Guido Freire, Joaquin Salvador Machulsky, Mario Leiva, Ana Lopez, Nicolas Spinelli, Luca Nicolás Forziati Gangi, Eitan Sprejer, Federico Barrera Lemarchand, Gerardo I. Simari, María Vanina Martínez · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28539 · cited by 0 works (OpenAlex) · open-access PDF

German

Wie können wir das Verhalten von AI-Supervize-Systemen, die die menschliche Intelligenz übertreffen, effektiv überwachen? AI-Sicherheit über Debatte schlägt vor, dass zwei AI-Systeme gegensätzliche Positionen festsetzen und ein Richter, entweder ein menschliches oder ein schwächeres Modell, entscheidet, welches das Richtige ist. Durch die Erleichterung von Supervisoren bei der Bereitstellung qualitativ hochwertiger AI-Trainingssignale zielt die Debatte darauf ab, das wahre und sichere Verhalten von AI-Systemen zu fördern, deren Fähigkeiten unsere eigenen übertreffen. Diese menschlichen Urteile sind jedoch nicht neutral: Menschen verlassen sich oft auf Vorurteile, die fortgeschrittene AI-Systeme ausnutzen können. Indem sie

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

Abstract

How can we effectively oversee AI systems that surpass human intelligence? AI Safety via Debate proposes that two AI systems argue opposing positions and a judge, either a human or a weaker model, decides which is correct. By facilitating supervisors in providing high-quality training signals, debate aims to encourage truthful and safe behavior from AI systems whose capabilities exceed our own. However, human judgment is not neutral: people often rely on biases which advanced AI systems may learn to exploit. Drawing on insights from collective intelligence and social cognition, this paper investigates whether debate can guide human and LLM supervisors toward the truth despite their prior beliefs. We evaluate several protocol configurations. We also explore setups inspired by the wisdom of the crowd such as the number and diversity of judges, as well as interaction mechanisms including deliberation and voting. Overall, we find that debate increases accuracy relative to prior beliefs for human judges, whereas the consultancy baseline, in which a single expert argues for one answer, does not. For weaker language models, however, the results are mixed: debate provides no comparable gains in most settings and can even reduce accuracy. Among multi-judge protocols, weighted voting produces the highest accuracy for human supervisors, whereas for LLMs none of them significantly improves group accuracy, even as the number of judges increases. Notably, the behavior of minority judges in non-diverse deliberation protocols suggests that, under symmetric settings, incorrect positions could be more persuasive than correct ones. These findings support debate as a viable path toward scalable human oversight, but cast doubt on its benefits for weaker AI supervisors, raising concerns about fully automated training and control pipelines in high-stakes settings.

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

Topics: Deliberation, Diversity (politics), Voting, Social psychology, Psychology, Computer science

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