arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2504.03255cs.CYcs.CLcs.MA
收藏

基于大语言模型的智能体系统中固有与涌现的责任问题:委托-代理视角

Inherent and emergent liability issues in LLM-based agentic systems: a principal-agent perspective

Queen Mary University of London(伦敦女王大学);UC Berkeley(伯克利大学);Certivize AI;UC San Francisco(旧金山大学)

Garry A. Gabison, R. Patrick Xian

AI总结:

本文从委托-代理视角分析LLM智能体系统的责任问题,涵盖部署中的责任后果,并提出可解释性、行为评估及故障安全机制等技术治理方向,以增强透明度和责任归属。

摘要:

由大语言模型(LLM)驱动的智能体系统正变得日益复杂和强大。其不断增强的自主性和不断扩展的部署场景,使得有效的治理政策、监控和控制协议日益受到关注。基于智能体市场的新兴格局,我们通过委托-代理视角,分析了委托使用LLM智能体及其扩展系统所产生的潜在责任问题。我们的分析补充了现有关于人工代理的风险研究,涵盖了委托-代理关系的重要方面及其在部署中的潜在后果。此外,我们推动了可解释性和行为评估、奖励与冲突管理,以及通过原则性的检测和故障安全机制工程来缓解错位与不当行为等技术治理方向的方法发展。通过阐明基于LLM的智能体系统中AI责任的突出问题,我们旨在为系统设计、审计和追踪提供信息,以增强透明度和责任归属。

英文摘要:

Agentic systems powered by large language models (LLMs) are becoming progressively more complex and capable. Their increasing agency and expanding deployment settings attract growing attention to effective governance policies, monitoring, and control protocols. Based on the emerging landscape of the agentic market, we analyze potential liability issues arising from the delegated use of LLM agents and their extended systems through a principal-agent perspective. Our analysis complements existing risk-based studies on artificial agency and covers the spectrum of important aspects of the principal-agent relationship and their potential consequences at deployment. Furthermore, we motivate method developments for technical governance along the directions of interpretability and behavior evaluations, reward and conflict management, and the mitigation of misalignment and misconduct through principled engineering of detection and fail-safe mechanisms. By illustrating the outstanding issues in AI liability for LLM-based agentic systems, we aim to inform the system design, auditing, and tracing to enhance transparency and liability attribution.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑