LLM-Powered Agents Simulate Workforce Responses to AI Integration
Researchers propose a computational testbed using LLM-powered generative agents to simulate employee psychological and behavioral responses to AI-driven workforce changes. The system combines management science and organizational behavior research with HR records, psychometric measures, and digital activity data to model cognitive, emotional, and behavioral trajectories. The architecture addresses privacy, accuracy, and representativeness concerns. The paper is published on arXiv under ID 2605.19064.
Key facts
- The testbed uses LLM-powered generative agents to simulate employee responses.
- It integrates management science and organizational behavior research.
- Agents are seeded with HR records, psychometric measures, and digital activity data.
- The simulation models cognitive, emotional, and behavioral trajectories across workdays.
- The paper details the computational architecture for the simulation platform.
- Privacy, accuracy, and representativeness standards are defined.
- The research is published on arXiv with ID 2605.19064.
- The testbed aims to forecast workforce transformations during AI integration.
Entities
Institutions
- arXiv