Owkin Launches Agentic AI Co-Pilot

- The K-Pro platform is underpinned by Owkin's proprietary federated learning technology, which allows AI models to be trained on decentralized datasets from different institutions without the data ever leaving its source, ensuring patient privacy. This architecture provides access to multimodal patient data, including genomics, imaging, and clinical records. - Owkin has secured over $320 million in funding from a mix of venture capital and strategic corporate investors, including Sanofi, Bristol-Myers Squibb, GV (formerly Google Ventures), and Bpifrance. This substantial backing includes a $180 million investment from Sanofi in 2021, which elevated Owkin to a "unicorn" status with a valuation over $1 billion. - The company's long-term vision is to create a "Biological Artificial Super Intelligence" (BASI), an AI system capable of modeling and engineering biology beyond human capabilities. K-Pro and the underlying OwkinZero biological large language model are positioned as key steps toward this goal. - In a real-world application, Owkin's platform has demonstrated the ability to reduce the time for drug target identification by 70%, from over a year to just three months. It has also been used to develop an IND-ready asset positioning strategy in a matter of hours. - K-Pro is designed with "agentic spaces," which are modular AI agents that can analyze data, activate insights, and propose next steps for tasks like clinical trial design and biomarker evaluation. This differs from general large language models by being grounded in specific biological reasoning tasks rather than abstract text generation. - Owkin has established significant collaborations with major pharmaceutical companies, including a multi-year, $90 million deal with Sanofi to discover and develop new treatments for four types of cancer. Additionally, a partnership with Bristol Myers Squibb, which included an $80 million investment, is focused on improving the design of clinical trials, initially in cardiovascular diseases. - To further advance its AI capabilities, Owkin announced a partnership with NVIDIA in January 2026 to scale its biological reasoning models using NVIDIA's NeMo framework and reinforcement learning tools. This collaboration aims to enhance the scalability and robustness of its agent-based systems. - The platform is not just for internal use; biopharma and academic clients can securely upload their own proprietary datasets to utilize K-Pro's analytical and reasoning capabilities through a natural language interface. Owkin also offers a free version, K-Pro FREE, for researchers to perform tasks like literature reviews and data visualization.