Research

Paper

TESTING February 18, 2026

Overseeing Agents Without Constant Oversight: Challenges and Opportunities

Authors

Madeleine Grunde-McLaughlin, Hussein Mozannar, Maya Murad, Jingya Chen, Saleema Amershi, Adam Fourney

Abstract

To enable human oversight, agentic AI systems often provide a trace of reasoning and action steps. Designing traces to have an informative, but not overwhelming, level of detail remains a critical challenge. In three user studies on a Computer User Agent, we investigate the utility of basic action traces for verification, explore three alternatives via design probes, and test a novel interface's impact on error finding in question-answering tasks. As expected, we find that current practices are cumbersome, limiting their efficacy. Conversely, our proposed design reduced the time participants spent finding errors. However, although participants reported higher levels of confidence in their decisions, their final accuracy was not meaningfully improved. To this end, our study surfaces challenges for human verification of agentic systems, including managing built-in assumptions, users' subjective and changing correctness criteria, and the shortcomings, yet importance, of communicating the agent's process.

Metadata

arXiv ID: 2602.16844
Provider: ARXIV
Primary Category: cs.HC
Published: 2026-02-18
Fetched: 2026-02-21 18:51

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.16844v1</id>\n    <title>Overseeing Agents Without Constant Oversight: Challenges and Opportunities</title>\n    <updated>2026-02-18T20:16:24Z</updated>\n    <link href='https://arxiv.org/abs/2602.16844v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.16844v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>To enable human oversight, agentic AI systems often provide a trace of reasoning and action steps. Designing traces to have an informative, but not overwhelming, level of detail remains a critical challenge. In three user studies on a Computer User Agent, we investigate the utility of basic action traces for verification, explore three alternatives via design probes, and test a novel interface's impact on error finding in question-answering tasks. As expected, we find that current practices are cumbersome, limiting their efficacy. Conversely, our proposed design reduced the time participants spent finding errors. However, although participants reported higher levels of confidence in their decisions, their final accuracy was not meaningfully improved. To this end, our study surfaces challenges for human verification of agentic systems, including managing built-in assumptions, users' subjective and changing correctness criteria, and the shortcomings, yet importance, of communicating the agent's process.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.HC'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-02-18T20:16:24Z</published>\n    <arxiv:primary_category term='cs.HC'/>\n    <author>\n      <name>Madeleine Grunde-McLaughlin</name>\n    </author>\n    <author>\n      <name>Hussein Mozannar</name>\n    </author>\n    <author>\n      <name>Maya Murad</name>\n    </author>\n    <author>\n      <name>Jingya Chen</name>\n    </author>\n    <author>\n      <name>Saleema Amershi</name>\n    </author>\n    <author>\n      <name>Adam Fourney</name>\n    </author>\n  </entry>"
}