Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gao, Ruijiang, Xiao, Steven Chong
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.16744
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914408154791936
author Gao, Ruijiang
Xiao, Steven Chong
author_facet Gao, Ruijiang
Xiao, Steven Chong
contents We study whether state-of-the-art AI coding agents, given the same data and research question, produce the same empirical results. Deploying 150 autonomous Claude Code agents to independently test six hypotheses about market quality trends in NYSE TAQ data for SPY (2015--2024), we find that AI agents exhibit sizable \textit{nonstandard errors} (NSEs), that is, uncertainty from agent-to-agent variation in analytical choices, analogous to those documented among human researchers. AI agents diverge substantially on measure choice (e.g., autocorrelation vs.\ variance ratio, dollar vs.\ share volume). Different model families (Sonnet 4.6 vs.\ Opus 4.6) exhibit stable ``empirical styles,'' reflecting systematic differences in methodological preferences. In a three-stage feedback protocol, AI peer review (written critiques) has minimal effect on dispersion, whereas exposure to top-rated exemplar papers reduces the interquartile range of estimates by 80--99\% within \textit{converging} measure families. Convergence occurs both through within-family estimation tightening and through agents switching measure families entirely, but convergence reflects imitation rather than understanding. These findings have implications for the growing use of AI in automated policy evaluation and empirical research.
format Preprint
id arxiv_https___arxiv_org_abs_2603_16744
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Nonstandard Errors in AI Agents
Gao, Ruijiang
Xiao, Steven Chong
Artificial Intelligence
Social and Information Networks
We study whether state-of-the-art AI coding agents, given the same data and research question, produce the same empirical results. Deploying 150 autonomous Claude Code agents to independently test six hypotheses about market quality trends in NYSE TAQ data for SPY (2015--2024), we find that AI agents exhibit sizable \textit{nonstandard errors} (NSEs), that is, uncertainty from agent-to-agent variation in analytical choices, analogous to those documented among human researchers. AI agents diverge substantially on measure choice (e.g., autocorrelation vs.\ variance ratio, dollar vs.\ share volume). Different model families (Sonnet 4.6 vs.\ Opus 4.6) exhibit stable ``empirical styles,'' reflecting systematic differences in methodological preferences. In a three-stage feedback protocol, AI peer review (written critiques) has minimal effect on dispersion, whereas exposure to top-rated exemplar papers reduces the interquartile range of estimates by 80--99\% within \textit{converging} measure families. Convergence occurs both through within-family estimation tightening and through agents switching measure families entirely, but convergence reflects imitation rather than understanding. These findings have implications for the growing use of AI in automated policy evaluation and empirical research.
title Nonstandard Errors in AI Agents
topic Artificial Intelligence
Social and Information Networks
url https://arxiv.org/abs/2603.16744