Anthropic is taking Claude beyond coding and document summaries, moving the AI system into drug discovery and scientific research. The company says it will launch its own preclinical programs while also introducing Claude Science, a new workbench designed for researchers and drugmakers.
A New Role for AI in Biotech
Announced at a San Francisco event on June 30, 2026, the initiative is built around neglected areas of medicine, including rare diseases. Anthropic says the goal is to support early-stage research before any treatment reaches human testing, using Claude to help explore molecules, biological pathways, and experimental strategies.
Company leaders describe the move as a practical test of their own technology. By working directly in the field, Anthropic aims to understand where AI can speed up scientific workflows and where human expertise remains essential. The company has not disclosed which conditions it will target or whether it plans to commercialize any future candidates.
Claude Science is presented as an AI workspace for scientific teams, with features tailored to genomics, proteomics, single-cell analysis, and cheminformatics. Anthropic says the system is built to keep results reproducible and traceable, a key requirement in research environments where precision matters.
Why Rare Diseases Matter
Anthropic is focusing on areas that larger drug pipelines often overlook. Rare diseases can offer a clearer biological starting point because many are linked to a single gene or a well-defined mechanism. That makes them a strong proving ground for AI-assisted discovery, even if the business model is more complex.
The company's approach also reflects a broader shift in science and technology: AI is no longer limited to analysis and automation, but is increasingly being used as a partner in discovery. Anthropic joins a growing group of organizations applying advanced models to biology, drug design, and research planning.
By entering its own programs, Anthropic can refine Claude for real scientific use while building closer feedback loops with researchers and industry partners. In the long run, this kind of integration could help shape faster, more adaptive tools for medicine and life sciences.