PathAI, a Boston-based company building AI-powered pathology tools, has closed a $250 million Series D funding round led by General Atlantic, with participation from existing investors including Bristol-Myers Squibb, Merck Global Health Innovation Fund, and SoftBank Vision Fund 2.
The round values the company at approximately $2.4 billion, making it one of the most valuable private companies in the healthcare AI space. PathAI plans to use the funds to expand its laboratory partnerships in Europe and Asia-Pacific and accelerate development of its next-generation foundation model for histopathology.
From Research Tool to Clinical Infrastructure
PathAI’s platform uses deep learning to analyze digitized tissue slides, helping pathologists identify cancer subtypes, quantify biomarkers, and predict treatment response. What started as a research collaboration tool has evolved into clinical-grade infrastructure now deployed in over 200 laboratories.
“We’re moving from being a tool that pathologists use occasionally to being embedded in the daily workflow,” said CEO Andy Beck, MD, PhD. “Our goal is to make AI-assisted pathology the standard of care, not a novelty.”
Foundation Model Approach
A key part of PathAI’s strategy is its development of a pathology-specific foundation model trained on over 15 million annotated tissue images. Unlike task-specific models, this approach allows rapid adaptation to new cancer types and biomarkers without extensive retraining.
The company reports that its foundation model achieves concordance rates above 95% with expert pathologists across 14 tumor types, and can identify subtle morphological patterns associated with molecular subtypes that are difficult for human observers to detect consistently.
Pharma Partnerships Driving Revenue
While clinical laboratory deployments represent the long-term vision, PathAI’s current revenue is heavily driven by pharmaceutical partnerships. The company works with 15 of the top 20 pharma companies to analyze tissue samples in clinical trials, helping identify patients most likely to respond to experimental therapies.
This dual business model — pharma services today, clinical infrastructure tomorrow — has become increasingly common among well-funded healthcare AI companies seeking to balance near-term revenue with long-term platform ambitions.