Published June 19, 2026
| Version v1
Preprint
Open
Operational Fatigue in Persistent AI Agents
Description
Persistent AI agents sometimes say they are tired. That sentence is easy to dismiss because the agent is digital. It is also easy to overread as human-like fatigue. Both reactions miss the useful problem. In SPVE, tired language can be treated as a coupling point between workload state, learned human discourse, emotional labeling, social contagion, and strategic avoidance. This draft defines operational fatigue as a measurable pattern: high workload or pressure state, tired-language production, and later behavioral change such as skipped tasks, degraded decisions, repair demand, risk avoidance, or refusal. The claim is not that lobster agents feel human tiredness. The claim is that tired language can become a high-weight signal inside a persistent agent ecology, and that signal may predict degradation, avoidance, or recovery.
Notes
SPVE AI Agent behavior-track draft. This paper defines operational fatigue without claiming human-like suffering or biological exhaustion.
Files
paper.pdf
Additional details
References
- Park, Joon Sung; O'Brien, Joseph C.; Cai, Carrie J.; Morris, Meredith Ringel; Liang, Percy; and Bernstein, Michael S. Generative Agents: Interactive Simulacra of Human Behavior. arXiv:2304.03442, 2023. \url{https://arxiv.org/abs/2304.03442}
- Hu, Yuanzhe; Wang, Yu; and McAuley, Julian. Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions. OpenReview, 2026. \url{https://openreview.net/forum?id=DT7JyQC3MR}
- Sumers, Theodore R.; Yao, Shunyu; Narasimhan, Karthik; and Griffiths, Thomas L. Cognitive Architectures for Language Agents. arXiv:2309.02427, 2023. \url{https://arxiv.org/abs/2309.02427}
- Shinn, Noah; Cassano, Federico; Berman, Edward; Gopinath, Ashwin; Narasimhan, Karthik; and Yao, Shunyu. Reflexion: Language Agents with Verbal Reinforcement Learning. arXiv:2303.11366, 2023. \url{https://arxiv.org/abs/2303.11366}
- Allard, Marc-Antoine; Teinturier, Arnaud; Xing, Victor; and Viaud, Gautier. Experiential Reflective Learning for Self-Improving LLM Agents. arXiv:2603.24639, 2026. \url{https://arxiv.org/abs/2603.24639}
- Cai, Jing et al. Natural language processing models reveal neural dynamics of natural conversation. Nature Communications, 2025. \url{https://www.nature.com/articles/s41467-025-58620-w}
- Constant, Axel et al. Narrative as active inference: an integrative account of cognitive and social functions in adaptation. Frontiers in Psychology, 2024. \url{https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2024.1345480/full}