A Stanford Medicine spinoff launched a “virtual biotech” that deploys thousands of AI scientist agents to execute parts of drug discovery without traditional lab infrastructure. The effort, created by biomedical data science associate professor James Zou and graduate student Harrison Zhang, reportedly runs the virtual lab model as an automated research workforce. The company is described as having 37,000 employees—an operational framing for AI agents rather than human staff—targeting workflow components that can be executed computationally. The model emphasizes scaling ideation, hypothesis generation, and experimentation planning while reducing constraints like lab space and human throughput. The development comes as the industry expands AI-native R&D operations, with increasing interest in how agentic systems can be validated against experimentally grounded biology. For biotech leaders, the approach is likely to raise questions about reproducibility, data provenance, and how quickly virtual findings can be transitioned into wet-lab execution under compliance constraints.