AI-assisted teams outperform AI-led teams but not human-only teams in assessing research reproducibility in quantitative social science

Question and contribution.

Can AI check whether published social science holds up? We randomly assigned 288 researchers to 103 teams that reproduced published results from leading social science journals working human-only, AI-assisted, or AI-led. Human-only and AI-assisted teams achieved comparable reproduction rates (94% vs. 91%), though human-only teams caught more major coding errors; AI-led teams reproduced only 37% and were weaker on every dimension. The design treats AI reliability as a measurable team-design question rather than a general claim about tools.

Abstract.

Large Language Models (LLMs) such as ChatGPT are transforming how scientists conduct and validate research, offering promise as tools to improve scientific reproducibility. However, computational reproducibility and error detection remain expensive and labor-intensive. We experimentally test how collaboration between researchers and LLM assistants influences the reproduction of quantitative social science findings across different levels of AI autonomy. We randomly assigned 288 researchers to 103 teams working under three conditions: human-only, AI-assisted (using ChatGPT as a collaborative tool), or AI-led (ChatGPT operating with minimal human oversight). Teams reproduced published results from leading social science journals, detected coding errors, and proposed robustness checks. Human-only and AI-assisted teams achieved comparable reproduction rates (94% vs. 91%) and performed similarly on most outcomes, except human-only teams identified significantly more major coding errors. Both substantially outperformed AI-led teams, which achieved only a 37% reproduction rate, detected fewer errors across all categories, proposed weaker robustness checks, and required more time. This autonomous approach, however, likely represents only a lower bound of AI capabilities. Despite rapid model advances, expert human judgment currently remains indispensable for reliable empirical verification. While AI assistance did not degrade most outcomes, it provided no measurable advantages and was associated with reduced detection of major errors. However, the 37% autonomous reproduction rate indicates that AI could provide value in settings where scale or cost constraints preclude human review of papers, even though general-purpose LLMs offer no immediate advantages for human-supervised verification.

How the project is framed.

BibTeX.

@article{aparicio2026ai,
  title = {AI-assisted teams outperform AI-led teams but not human-only teams in assessing research reproducibility in quantitative social science},
  author = {Abel Brodeur and David Valenta and Alexandru Marcoci and Juan P. Aparicio and Derek Mikola and Bruno Barbarioli and others},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2026},
  doi = {10.1073/pnas.2524747123}
}