mlsanityData to 13 Oct 2026

Stochastic Parrots or Singing in Harmony? Testing Five Leading LLMs for their Ability to Replicate a Human Survey with Synthetic Data

arXiv 2603.00059 ↗First posted 10 Feb 2026cs.CY cs.AIcited by 2

On the map: Synthetic data and self-improvement for LLMs → LLM synthetic data generation

Abstract (arXiv)

How well can AI-derived synthetic research data replicate the responses of human participants? An emerging literature has begun to engage with this question, which carries deep implications for organizational research practice. This article presents a comparison between a human-respondent survey of 420 Silicon Valley coders and developers and synthetic survey data designed to simulate real survey takers generated by five leading Generative AI Large Language Models: ChatGPT Thinking 5 Pro, Claude Sonnet 4.5 Pro plus Claude CoWork 1.123, Gemini Advanced 2.5 Pro, Incredible 1.0, and DeepSeek 3.2. Our findings reveal that while AI agents produced technically plausible results that lean more towards replicability and harmonization than assumed, none were able to capture the counterintuitive insights that made the human survey valuable. Moreover, deviations grouped together for all models, leaving the real data as the outlier. Our key finding is that while leading LLMs are increasingly being used to scale, replicate and replace human survey responses in research, these advances only show an increased capacity to parrot conventional wisdom in harmony with each other rather than revealing novel findings. If synthetic respondents are used in future research, we need more replicable validation protocols and reporting standards for when and where synthetic survey data can be used responsibly, a gap that this paper fills. Our results suggest that synthetic survey responses cannot meaningfully model real human social beliefs within organizations, particularly in contexts lacking previously documented evidence. We conclude that synthetic survey-based research should be cast not as a substitute for rigorous survey methods, but as an increasingly reliable pre- or post-fieldwork instrument for identifying societal assumptions, conventional wisdoms, and other expectations about research populations.

Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Placed in LLM synthetic data generation, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 17 of its 72 references that are in the map. First posted on arXiv 10 Feb 2026.

Builds on

17 of 72 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Synthetic Data Generation Using Large Language Models: Advances in Text and Code18 Mar

Other references · 16

  1. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,541
  2. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 747
  3. 2023Bias and Fairness in Large Language Models: A Surveycited by 636
  4. 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 512
  5. 2023Prediction-Powered Inferencecited by 218
  6. 2022Thinking Fast and Slow in Large Language Modelscited by 136
  7. 2022Meaning without reference in large language modelscited by 76
  8. 2023Using Imperfect Surrogates for Downstream Inference: Design-based Supervised Learning for Social Science Applications of Large Language Modelscited by 38
  9. 2022Synthetic Data in Human Analysis: A Surveycited by 28
  10. 2024Vox Populi, Vox AI? Using Language Models to Estimate German Public Opinioncited by 13
  11. 2023Can AI Serve as a Substitute for Human Subjects in Software Engineering Research?cited by 8
  12. 2025AIn't Nothing But a Survey? Using Large Language Models for Coding German Open-Ended Survey Responses on Survey Motivationcited by 4
4 more
  1. 2025Bridging Information Gaps with Comprehensive Answers: Improving the Diversity and Informativeness of Follow-Up Questionscited by 2
  2. 2025Evaluating the Use of Large Language Models as Synthetic Social Agents in Social Science Researchcited by 2
  3. 2025A New Digital Divide? Coder Worldviews, the Slop Economy, and Democracy in the Age of AIcited by 2
  4. 2023Fully Synthetic Data for Complex Surveyscited by 1

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026The conditional superiority of fast silicon samplingcited by 1
  2. 2026Shared Autoregressive Context Can Distort Relationships in Synthetic Datacited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Evaluating the Bias in LLMs for Surveying Opinion and Decision Making in Healthcare
  2. 2025Are LLM Agents Behaviorally Coherent? Latent Profiles for Social Simulation
  3. 2026Towards Detecting AI-Assisted Responses in Online Surveys
  4. 2024LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individuals
  5. 2025Emulating Public Opinion: A Proof-of-Concept of AI-Generated Synthetic Survey Responses for the Chilean Case
  6. 2026Evaluating LLMs as Human Surrogates in Controlled Experiments
  7. 2025Can Generative AI agents behave like humans? Evidence from laboratory market experiments
  8. 2026Synthetic Data in Marketing Research: How to Evaluate and When to Trust