Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems
Summary
This position paper argues that AI agents in scientific teams should be studied as human-agent systems to enhance collaboration and mitigate risks such as reduced diversity in scientific inquiry.
View Cached Full Text
Cached at: 08/18/26, 03:51 AM
Paper page - Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems
Source: https://huggingface.co/papers/2608.14667 Published on Aug 2
·
Submitted byhttps://huggingface.co/LeoYML
Leo Yon Aug 18
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
Scientific collaboration with AI agents requires studying human-agent pairs to avoid risks like reduced inquiry diversity and to foster synergistic discovery.
Large language model-based agentsare increasingly deployed as collaborators in scientific discovery yet most current work focuses on the autonomous capabilities of “AI Scientists”. We argue that this overlooks the social aspects of scientific teamwork, and that studyingAI Scientistsashuman-agent systems(HAS)--where the unit of analysis is the human-agent pair--is both underexplored and undervalued. We establish these points through literature and empirical analysis, and highlight recent incidences and studies which show that deploying agents in science without accounting forhuman-agent dynamicsintroduces near-term risks, including reduced diversity of scientific inquiry. Through analysis of real-world case studies, we show that scientists and agents can augment each other’s capabilities. We call for new research that adopts the HAS lens to develop mathematical frameworks for understanding and fosteringhuman-AI synergyin scientific discovery.
View arXiv pageView PDFAdd to collection
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2608.14667 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2608.14667 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2608.14667 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
AI Evaluation Should Work With Humans
This position paper argues that AI evaluation should pivot to assessing human-AI teams rather than superhuman performance to foster better societal outcomes.
AI Agents Push Humans Out of the Loop
The paper argues that current AI agent systems degrade human oversight capabilities and advocates for design-level changes to support effective human-agent interaction, emphasizing the need to maintain human skills.
Position: Behavioral Systems Require Behavioral Tests
This position paper argues that AI agents must be evaluated like behavioral systems through systematic observation and testing of their processes, not just performance outcomes, and proposes a research agenda to develop rigorous behavioral tests for better understanding and safety.
Agentic AI for Safety-critical Multi-drone Systems: Challenges and Opportunities
This position paper discusses the challenges and opportunities of integrating agentic AI into safety-critical multi-drone systems, advocating for human-centered, socio-technical design approaches to ensure trust, governance, and adoption in professional contexts.
@JIACHENLIU8: AI4S research today: self-evolving loops, multi-agent systems, agent skills, science benchmarks etc. --> all to make li…
This article critiques current AI for Science (AI4S) research for focusing on compensating weaknesses of frontier models, arguing that the true bottleneck is the scientific ecosystem rather than individual AI scientist intelligence. It calls for a shift toward building an AI-native scientific ecosystem from first principles.