Owkin

Owkin is a pioneering AI biotech company and enterprise-grade federated learning platform that accelerates drug discovery, develops FDA-cleared AI diagnostics, and optimizes clinical trials through privacy-preserving machine learning on real-world hospital data, with all access provided through custom enterprise partnerships priced via negotiated contracts.

Owkin has built its entire platform around federated learning—a privacy-preserving approach that trains machine learning models directly on hospital data without ever centralizing sensitive patient records. This architectural choice distinguishes Owkin from competitors like Recursion Pharmaceuticals, Insilico Medicine, and BenevolentAI, which rely primarily on centralized datasets or proprietary lab-generated data. Founded in 2016 and headquartered in Paris, the company has raised over $300M in funding, maintains offices across seven locations (Paris, London, New York, Cambridge MA, Geneva, Nantes, and additional sites), and has published more than 80 peer-reviewed papers in journals including Nature Medicine and The Lancet Digital Health.

How Owkin Works

At its core, Owkin's MOSAIC platform deploys machine learning models to 15+ leading academic medical centers worldwide. Rather than transferring sensitive patient data to a central server, MOSAIC sends algorithms to the data. Each hospital trains the model locally on its own pathology images, genomic profiles, and clinical records, then returns only the learned model parameters—never the raw patient data—back to the central system for aggregation. This federated architecture ensures full GDPR and HIPAA compliance while enabling Owkin to train on some of the largest and most diverse real-world clinical datasets available.

The company's computational pathology engine applies deep learning and computer vision to whole-slide H&E-stained pathology images, extracting biological features invisible to the human eye that correlate with clinical outcomes, genomic profiles, and treatment responses. This capability powers both Owkin's diagnostic products and its biomarker discovery pipeline.

Key Products and Partnerships

Owkin's most commercially mature product is MSIntuit CRC, an FDA 510(k)-cleared AI diagnostic that detects microsatellite instability (MSI) in colorectal cancer directly from standard pathology slides. MSIntuit CRC can triage approximately 30% of colorectal cancer cases as microsatellite stable (MSS) without requiring confirmatory molecular testing, reducing both turnaround time and lab costs.

The K Pro AI Scientist is Owkin's pharma decision-making tool, synthesizing multimodal data—including spatial and multi-omics reporting—to help pharmaceutical R&D leaders prioritize assets, design trials, and select patient populations. K Pro supports three decision categories: clinical trial decisions, patient and population decisions, and early portfolio decisions.

Owkin's strategic partnerships with Sanofi (valued at over $180M, announced 2021) and Bristol Myers Squibb validate the platform's value for top-tier pharmaceutical companies. These collaborations span oncology drug discovery, predictive biomarker development, and clinical trial optimization across tumor types including non-small cell lung cancer, mesothelioma, breast cancer, and hepatocellular carcinoma.

Who Owkin Is For

Owkin is designed for large pharmaceutical companies, academic medical centers, and hospital networks pursuing AI-driven oncology research at scale. The platform is not available on a self-service basis and requires multi-million-dollar enterprise partnerships. Individual researchers, small biotech startups, and clinical practitioners without institutional backing will find the platform inaccessible. Prospective clients should expect a full sales cycle with no publicly available pricing, and should benchmark Owkin's federated learning approach against the centralized or generative AI models offered by Recursion, Insilico Medicine, and PathAI.