Anthropic is building a biology laboratory in the San Francisco area where Claude will guide robots through physical experiments. The project moves the company’s life-science work beyond software analysis and into automated laboratory workflows, with human oversight retained for safety.
The company has already introduced Claude Science, a workspace for researchers, and a Model Hardware Standard for controlling laboratory equipment. Together, those tools are intended to let the model plan and coordinate experiments while robots handle repeatable physical steps.
Anthropic’s life-sciences head Eric Kauderer-Abrams sees a potential use in designing complex antibodies for diseases considered difficult to target with existing drugs. The company strengthened its biology operation in April by acquiring Coefficient Bio for about $400 million and added Novartis chief executive Vas Narasimhan to its board.
Anthropic does not plan to conduct its own clinical trials at this stage, according to the report. The lab can test whether AI-directed experimentation speeds early research, but successful laboratory results would still face independent validation, safety studies and the long regulatory process required before any treatment reaches patients.