Stanford researchers unveiled Biomni, an AI “co-scientist” designed to autonomously run biomedical research workflows rather than only answer questions. The system is built to interpret natural language research prompts, generate hypotheses, select datasets, and write code that supports experimental planning. The announcement positions Biomni as an automation layer for parts of the discovery pipeline that traditionally require manual orchestration across literature review, analysis, and protocol drafting. The work also links Biomni’s design to Stanford’s broader AI research ecosystem, including a spinout collaboration. For biotech teams, the core question becomes how quickly agentic tools can reduce cycle times in hypothesis-to-experiment planning without degrading scientific quality controls that typically require expert oversight.