发表机构
Cornell Tech / Cornell University; University of Maryland, College Park; Korea Advanced Institute of Science and Technology（康奈尔科技校区 / 康奈尔大学; 马里兰大学帕克分校; 韩国科学技术院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出运营韧性与体贴参与两个评估维度，通过模拟医疗场景实验，揭示智能体在累积挑战下行为变化，并推导出五个部署困境及技术启示。
AI 中文摘要
生成式人工智能智能体的持续部署需要的不仅仅是孤立的任务成功。智能体必须在重复交互、变化条件以及对共享工作流中人员的依赖中保持有用，尤其是在技术、人力和运营干扰随时间累积的情况下。我们提出将运营韧性和体贴参与作为评估此类智能体的两个互补方面：前者捕捉智能体如何从受阻工作中恢复，同时保留进展并传达其局限性；后者捕捉其适应如何考虑到受影响的人员、角色边界和周围的工作流程。然而，在累积挑战下，这两方面仍未得到充分探索。我们研究了在轻度、中度和重度挑战下，两种生成式人工智能模型和十二个利益相关者衍生任务中的120条模拟医疗轨迹。我们比较了文本行动计划、提示式内部评估以及定量结构化工作量和情感报告，以考察智能体行为和报告状态如何随挑战累积而变化。关于运营韧性，智能体从自我导向的恢复转向更大的人类依赖，同时在结构化报告中报告工作量和负面情感增加，但在文本响应中很少表达压力。关于体贴参与，智能体从以任务为中心的适应扩展到任务重构、关注他人、角色边界调整和更广泛的协调，在行动和内部评估中表现出不同的模式。从这些发现中，我们推导出五个部署困境，涉及坚持、注意力、角色边界、状态披露和升级，这些需要利益相关者的明确说明，进一步为学习、情境评估和具身适应的技术启示提供信息。
英文摘要
Sustained deployment of generative AI agents requires more than isolated task success. Agents must remain useful across repeated interactions, changing conditions, and dependencies on people within shared workflows, especially as technical, human, and operational disruptions accumulate over time. We propose operational resilience and considerate participation as two complementary aspects of evaluating such agents: the former captures how agents recover from blocked work while preserving progress and communicating their limits, and the latter captures how their adaptation accounts for affected people, role boundaries, and the surrounding workflow. Yet both remain underexplored under accumulating challenge. We study 120 simulated healthcare trajectories across two generative AI models and twelve stakeholder-derived tasks under light, medium, and heavy challenge. We compare textual action plans, prompted internal assessments, and quantitative structured workload and affect reports to examine how agent behavior and reported state change as challenge accumulates. Regarding operational resilience, agents shift from self-directed recovery toward greater human dependence, while reporting increasing workload and negative affect in structured reports but seldom expressing strain in textual responses. Regarding considerate participation, agents broaden from task-focused adaptation toward task reframing, attention to others, role-boundary adjustment, and wider coordination, with distinct patterns across actions and internal assessments. From these findings, we derive five deployment dilemmas involving persistence, attention, role boundaries, state disclosure, and escalation that require stakeholder specification, further informing technical implications for learning, situated evaluation, and embodied adaptation.
