Anthropic is building its own drug discovery programs to develop treatments for diseases that traditional pharmaceutical companies ignore because they lack profit potential. The AI company targets neglected diseases, a category that includes tropical infections and rare conditions affecting primarily low-income populations.
The move represents a direct application of Anthropic's large language models to biomedical research. Rather than licensing technology to existing pharma companies, Anthropic takes on the development burden itself, positioning AI as a tool to make economically unviable research feasible.
Novartis CEO Vas Narasimhan outlined the potential impact. AI could compress drug development timelines from twelve years to seven or eight years while boosting success rates from the industry standard of 8 percent to 16 percent. These improvements address two core constraints that make neglected disease research unattractive to Big Pharma. Shorter timelines reduce capital burn. Higher success rates improve return on investment.
Anthropic's strategy acknowledges that market forces alone won't solve global health gaps. Companies pursuing maximum profit naturally avoid diseases affecting populations with limited purchasing power. Orphan drugs and neglected tropical diseases remain underfunded despite their health burden in developing nations.
The company hasn't disclosed specific diseases it will target or timelines for drug candidates. Drug discovery remains capital-intensive and scientifically uncertain even with AI assistance. Developing a single approved medication typically costs over $2 billion and requires passage through multiple regulatory phases.
Anthropic's entry into drug discovery also reflects broader ambitions for Claude, its flagship model. The company has been methodically expanding from chatbot applications into specialized domains. Biomedical research demands reasoning about complex molecular interactions and literature synthesis, tasks where large language models show promise.
This initiative tests whether AI can solve coordination and efficiency problems that plague neglected disease research. Success would validate using AI not just to optimize profitable sectors but to address market