Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluations

ACL ARR 2026 January Submission8161 Authors

Keywords: Evaluation, LLM Agents, Validity, Human-Centered NLP, Bias and Fairness, User Study

Abstract: Agentic benchmarks increasingly rely on LLM-simulated users to scalably evaluate agent performance, yet the robustness, validity, and fairness of this approach remain unexamined. Through a user study with participants across the United States, India, Kenya, and Nigeria, we investigate whether LLM-simulated users serve as reliable proxies for real human users in evaluating agents on τ-Bench retail tasks. We find that user simulators lack robustness, with agent success rates varying up to 9 percentage points across different user LLMs. Furthermore, simulated users systematically miscalibrate performance, underestimating success on challenging tasks while overestimating moderately difficult ones. African American Vernacular English (AAVE) speakers experience consistently worse success rates and calibration errors than Standard American English (SAE) speakers, with disparities compounding significantly with age. We also find simulated users to be a differentially effective proxy for different populations, performing worst for AAVE and Indian English. Additionally, simulated users introduce conversational artifacts and surface different failure patterns than human users. These findings demonstrate that current evaluation practices risk misrepresenting agent capabilities across diverse user populations and may obscure real-world deployment challenges.

Paper Type: Long

Research Area: Human-AI Interaction/Cooperation and Human-Centric NLP

Research Area Keywords: human-centered evaluation, human-AI interaction

Contribution Types: Model analysis & interpretability, Data analysis

Languages Studied: English

Submission Number: 8161