Research

Paper

AI LLM March 17, 2026

Nonstandard Errors in AI Agents

Authors

Ruijiang Gao, Steven Chong Xiao

Abstract

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.

Metadata

arXiv ID: 2603.16744
Provider: ARXIV
Primary Category: cs.AI
Published: 2026-03-17
Fetched: 2026-03-18 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.16744v1</id>\n    <title>Nonstandard Errors in AI Agents</title>\n    <updated>2026-03-17T16:21:22Z</updated>\n    <link href='https://arxiv.org/abs/2603.16744v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.16744v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SI'/>\n    <published>2026-03-17T16:21:22Z</published>\n    <arxiv:comment>45 pages</arxiv:comment>\n    <arxiv:primary_category term='cs.AI'/>\n    <author>\n      <name>Ruijiang Gao</name>\n    </author>\n    <author>\n      <name>Steven Chong Xiao</name>\n    </author>\n  </entry>"
}