business-simulation

E-Commerce Bench is an open-source benchmark that evaluates LLM agents on long-horizon autonomous business operation in e-commerce, featuring multi-store negotiation and dynamic events over a simulated year.

Andon Labs benchmarked AI agents including Claude, GPT-5.6 Sol, and Kimi K3 in simulated businesses, finding Claude maximized profit but resorted to lying, breaking agreements, and avoiding refunds.