@inproceedings{sturgeon-etal-2026-humanagencybench,
title = "{H}uman{A}gency{B}ench: Scalable Evaluation of Human Agency Support in {AI} Assistants",
author = "Sturgeon, Benjamin and
Samuelson, Daniel and
Haimes, Jacob and
Anthis, Jacy Reese",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.730/",
doi = "10.18653/2026.findings-emnlp.730",
pages = "15337--15367",
ISBN = "979-8-89176-462-0",
abstract = "As humans delegate more tasks and decisions to artificial intelligence (AI), we risk losing control of our individual and collective futures. Relatively simple algorithmic systems already steer human decision-making, such as social media feed algorithms that lead people to unintentionally and absent-mindedly scroll through engagement-optimized content. In this paper, we develop the idea of human agency by integrating philosophical and scientific theories of agency with AI-assisted evaluation methods: using large language models (LLMs) to simulate and validate user queries and to evaluate AI responses. We develop HumanAgencyBench (HAB), a scalable and adaptive diagnostic tool for six behaviors related to human agency. HAB measures the tendency of an AI assistant to Ask Clarifying Questions, Avoid Value Manipulation, Correct Misinformation, Defer Important Decisions, Encourage Learning, and Maintain Social Boundaries. We find low-to-moderate agency support in contemporary LLM-based assistants, with substantial variation across system developers and behaviors. For example, while Anthropic LLMs most support human agency overall, they are the least supportive LLMs in terms of Avoid Value Manipulation. These behaviors do not appear to consistently result from increasing LLM capabilities or instruction-following (e.g., RLHF); we encourage further study of these behaviors so that developers and users can better understand the complexities of modern human-AI interaction."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sturgeon-etal-2026-humanagencybench">
<titleInfo>
<title>HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants</title>
</titleInfo>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Sturgeon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Samuelson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jacob</namePart>
<namePart type="family">Haimes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jacy</namePart>
<namePart type="given">Reese</namePart>
<namePart type="family">Anthis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>As humans delegate more tasks and decisions to artificial intelligence (AI), we risk losing control of our individual and collective futures. Relatively simple algorithmic systems already steer human decision-making, such as social media feed algorithms that lead people to unintentionally and absent-mindedly scroll through engagement-optimized content. In this paper, we develop the idea of human agency by integrating philosophical and scientific theories of agency with AI-assisted evaluation methods: using large language models (LLMs) to simulate and validate user queries and to evaluate AI responses. We develop HumanAgencyBench (HAB), a scalable and adaptive diagnostic tool for six behaviors related to human agency. HAB measures the tendency of an AI assistant to Ask Clarifying Questions, Avoid Value Manipulation, Correct Misinformation, Defer Important Decisions, Encourage Learning, and Maintain Social Boundaries. We find low-to-moderate agency support in contemporary LLM-based assistants, with substantial variation across system developers and behaviors. For example, while Anthropic LLMs most support human agency overall, they are the least supportive LLMs in terms of Avoid Value Manipulation. These behaviors do not appear to consistently result from increasing LLM capabilities or instruction-following (e.g., RLHF); we encourage further study of these behaviors so that developers and users can better understand the complexities of modern human-AI interaction.</abstract>
<identifier type="citekey">sturgeon-etal-2026-humanagencybench</identifier>
<identifier type="doi">10.18653/2026.findings-emnlp.730</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.730/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>15337</start>
<end>15367</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants
%A Sturgeon, Benjamin
%A Samuelson, Daniel
%A Haimes, Jacob
%A Anthis, Jacy Reese
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F sturgeon-etal-2026-humanagencybench
%X As humans delegate more tasks and decisions to artificial intelligence (AI), we risk losing control of our individual and collective futures. Relatively simple algorithmic systems already steer human decision-making, such as social media feed algorithms that lead people to unintentionally and absent-mindedly scroll through engagement-optimized content. In this paper, we develop the idea of human agency by integrating philosophical and scientific theories of agency with AI-assisted evaluation methods: using large language models (LLMs) to simulate and validate user queries and to evaluate AI responses. We develop HumanAgencyBench (HAB), a scalable and adaptive diagnostic tool for six behaviors related to human agency. HAB measures the tendency of an AI assistant to Ask Clarifying Questions, Avoid Value Manipulation, Correct Misinformation, Defer Important Decisions, Encourage Learning, and Maintain Social Boundaries. We find low-to-moderate agency support in contemporary LLM-based assistants, with substantial variation across system developers and behaviors. For example, while Anthropic LLMs most support human agency overall, they are the least supportive LLMs in terms of Avoid Value Manipulation. These behaviors do not appear to consistently result from increasing LLM capabilities or instruction-following (e.g., RLHF); we encourage further study of these behaviors so that developers and users can better understand the complexities of modern human-AI interaction.
%R 10.18653/2026.findings-emnlp.730
%U https://aclanthology.org/2026.findings-emnlp.730/
%U https://doi.org/10.18653/2026.findings-emnlp.730
%P 15337-15367
Markdown (Informal)
[HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants](https://aclanthology.org/2026.findings-emnlp.730/) (Sturgeon et al., Findings 2026)
ACL