Automating and Scaling Behavioral Scientific Research on AI Agents

[Submitted on 10 Aug 2026]

View PDF HTML (experimental)

Abstract:As AI agents are increasingly deployed in complex environments, understanding their behaviors becomes critical. Yet behavioral scientific research on AI agents remains manual and labor-intensive. We introduce AEROBAT, the first multi-agent system to automate behavioral scientific research on AI agents. Given an arbitrary target behavior by its user, AEROBAT automatically executes a full pipeline of behavioral scientific research—generating hypotheses about the behavior, designing and executing controlled experiments, making behavioral assessments, analyzing the results, and writing reports. For 12 target behaviors, we used AEROBAT to generate and test 79 hypotheses: designing 1,240 controlled experiments and executing 23,512 simulation rounds in total. Moderate-to-strong statistical evidence was found for 26 hypotheses, including some novel ones. In sum, our results demonstrate that automated behavioral scientific research on AI agents can complement and extend the reach of manual research.

Submission history

From: Soo Yong Lee [view email]
[v1] Mon, 10 Aug 2026 01:14:31 UTC (1,702 KB)