RESEARCHInvestigateNEXT 12 MONTHS
LLMs struggle to simulate human belief updates in controlled environments
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Research shows six leading LLMs fail to accurately replicate human belief and stance updates when compared directly to human participants.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 4 November 2026
- Collected by OneBench
- 31 Jul 2026, 09:36 UK
Stored source excerpt
arXiv:2607.28347v1 Announce Type: new Abstract: LLMs are increasingly deployed as proxies for human study participants in social science experiments, yet the fidelity of this practice…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.