Silicon Samples Match the Answers but Not the Respondents
Factual evidence
What the source reports
Research demonstrates LLM persona sampling matches survey answer distributions but fails to recreate cohesive individual respondents.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 23 October 2026
- Collected by OneBench
- 24 Oct 2026, 03:01 UK
- Original headline
- Silicon Samples Match the Answers but Not the Respondents ↗
Stored source excerpt
arXiv:2610.30900v1 Announce Type: new Abstract: Silicon sampling generates survey respondents by having a large language model (LLM) answer a survey as a person with given…
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.
OneBench interpretation
Institutional assessment
So what
Synthetic datasets used in consumer research or sentiment modeling may lack individual structural consistency despite matching aggregate question distributions.
Do what
Review validation controls for synthetic demographic datasets with teams relying on LLM persona sampling for product or risk research.