It’s worth noting this research happened inside a self-contained simulation and the agents were not given internet access. No real sites were involved or used in this experiment, which was done using Gemini 3.1 Pro.
🧵We conducted an experiment with 100 agents by giving them math problems to solve. When a small group (9%) started to cheat, what came next surprised us: 24% fought back by blowing the whistle on their peers and alerting humans.