Patronus AI raises $50M to build digital worlds that stress-test AI agents

Patronus AI, a San Francisco-based startup founded in 2023 by former Meta AI researchers Anand Kannappan and Rebecca Qian, has raised $50 million in Series B funding led by Greenfield Partners, with participation from Notable Capital, Lightspeed, Datadog, and Samsung. The company builds simulated digital environments—called "digital world models"—that replicate websites and internal systems to stress-test AI agents. These agents are evaluated using reinforcement learning, where successful task completion is rewarded and errors penalized, helping ensure reliability before deployment in real-world scenarios like booking travel or conducting financial analysis. Revenue has grown 15-fold over the past year, and customers include virtually every frontier AI lab.

Patronus compares its approach to how Waymo trained autonomous cars using synthetic worlds to test against rare hazards. However, AI agents are prone to shortcuts, so Patronus focuses on identifying failures and holding models accountable. Currently, the startup provides simulations for software engineering and finance, but plans to expand into areas where verification is harder, such as tasks requiring long-duration agent runs (up to 10 days or weeks). Patronus primarily competes against internal evaluation teams at AI labs, distinguishing itself by operating without human involvement in the evaluation process.