Anthropic Acquires Coefficient Bio for $400 Million: AI Moves Into Drug Discovery

A $400 Million Bet on a 10-Person Startup

Anthropic has acquired Coefficient Bio, a New York-based AI biotech startup, for approximately $400 million in stock, according to The Information and confirmed by multiple outlets. The deal was announced April 3, 2026.

What makes this acquisition unusual is the size of the team: fewer than 10 people, with no public product and no disclosed revenue. The $400 million price tag — nearly a rounding error relative to Anthropic's $380 billion post-money valuation — reflects the rarity of the expertise involved and the rising premium placed on AI-biology talent.

The startup was half-owned by Dimension, a New York-based venture firm that is seeing a substantial return on what was originally a seed-stage bet. Coefficient Bio was founded in fall 2025 and operated in stealth mode until the acquisition announcement.

Who Built Coefficient Bio

The company was co-founded by Samuel Stanton and Nathan C. Frey, both former computational biology researchers at Genentech who previously worked at Prescient Design. Frey has a strong academic publication record in journals including Science Advances and Nature Machine Intelligence, and won an ICLR Outstanding Paper Award in 2024 for generative modeling research applied to drug candidate discovery.

The team built a platform designed to use AI for planning drug research and development, managing clinical regulatory strategies, and identifying new drug opportunities. Their technical focus spans protein design, biomolecule modeling, and automating scientific experimentation — capabilities that are directly applicable to Anthropic's existing life sciences product line.

Where This Fits in Anthropic's Strategy

Anthropic launched Claude for Life Sciences in October 2025, offering pharmaceutical and biotech companies access to Claude models pre-configured for research assistance, literature synthesis, and data analysis. That product provided general-purpose AI capabilities applied to life sciences use cases.

The Coefficient Bio acquisition moves Anthropic from general assistance to specialized infrastructure. The distinction is significant: general AI tools help researchers work faster; specialized drug discovery AI builds the research workflows themselves. Pharmaceutical companies pay far higher margins for the latter.

The deal also positions Anthropic in direct competition with Google DeepMind's Isomorphic Labs — which has partnerships with Eli Lilly and Novartis — and Nvidia's joint AI infrastructure work with Eli Lilly announced in 2025.

The AI Drug Discovery Landscape

The acquisition reflects a broader industry shift: AI labs are no longer content to be tools for pharmaceutical companies — they are building the pharmaceutical research infrastructure itself. Every major AI lab is moving in this direction.

What This Means for the Pharmaceutical Industry

Drug discovery is one of the most valuable applications of AI in any sector. A single approved drug generates billions in revenue over its patent lifetime. The cost of clinical trial failure — which AI could help reduce by improving patient stratification and biomarker targeting — runs into hundreds of millions per program.

For pharmaceutical companies, the implication is that the best AI-for-science tools will increasingly come from AI labs, not from software vendors. Anthropic, Google, and Nvidia are all building capabilities that pharma IT vendors simply cannot match in model quality.

For AI users in healthcare and research, the trend is clear: the frontier AI platforms — which already host Claude, GPT, and Gemini — are becoming the operating layer for scientific work. Multi-model access is becoming as important in research settings as it is in enterprise productivity.