Insitro

Insitro is an AI-native biotech that combines 20+ petabytes of automated cellular experiments with population-scale genetics to build "Physical AI" systems that reveal the true causal drivers of human disease. Founded by Daphne Koller, the company targets a fundamental problem in drug discovery: most targets fail because they aren't truly causal in human disease. By integrating multimodal clinical and cellular data, insitro aims to identify high-confidence targets and design better medicines. The platform centers on two engines. Virtual Human is a genetically anchored causal AI engine built on the world's largest integrated multimodal corpus of human clinical and cellular data. It reveals how disease begins and progresses, pinpointing causal targets before candidate optimization. TherML is a modality-agnostic therapeutic design engine spanning small molecules, oligonucleotides, and complex biologics. Following the 2026 acquisition of CombinAbleAI, TherML now unifies design across all these modalities, matching the best-fit drug to each target. Recent progress validates the approach. At ADA 2026, insitro presented preclinical data showing its AI-discovered MASH candidate reduces fibrosis independent of liver-fat reduction. The company expanded its ALS collaboration with BMS, nominating two additional targets from Virtual Human. In a landmark AI-enabled human genetics study of brown adipose tissue, insitro identified anti-obesity targets, with BAT-01 knockdown achieving 15% body weight reduction in preclinical models. Insitro is not a software product; it's a partnership- and pipeline-driven biotech. Value is created through internal programs and collaborations with pharma. For investors and partners, the momentum is real—BMS expansion, MASH data, CombinAbleAI acquisition. For teams seeking a plug-and-play AI tool, this isn't it. Insitro is a bet on a data-dense, causal-AI approach to drug discovery, not a piece of software you can buy.