Owkin

by Owkin  · Based in France → — Building Biological Artificial Superintelligence
genetics Oncology Pathology
Website Demo Contact
Claim this tool
Is this the maker? Claim the company →

Regulatory Status Disclosed

Overview

Owkin is at the forefront of developing AI solutions for drug discovery and real-world evidence, leveraging a unique approach called federated learning. This technology allows AI models to be trained across decentralized hospital networks without centralizing sensitive patient data, ensuring privacy and security while maximizing data utility. Their core offering, K Pro, is an AI Scientist designed to accelerate biomedical research and generate clinical insights.

K Pro functions as an AI agent that connects research and patient care. It continuously learns from real-world data, user feedback, and clinical validation, evolving towards fully automated R&D. For physicians, this means access to advanced insights that can inform clinical trial decisions, patient and population decisions, and early portfolio decisions. The platform is capable of generating bespoke reports powered by spatial biology and multi-omics data, offering a deeper understanding of disease mechanisms.

Owkin’s vision is to create an autonomous AI Scientist capable of true causal understanding of disease, ultimately leading to a world where R&D is automated and research is directly connected to care. By pioneering advanced AI that can explore and formulate hypotheses beyond human limits, Owkin aims to tackle the profound complexity of biology and accelerate the development of new treatments.

Reviewed by Pouyan Golshani, MD — Interventional Radiologist

Key Features

  • Federated learning for data privacy
  • AI Scientist (K Pro) for R&D automation
  • Multimodal patient data network
  • Spatial biology and multi-omics reports
  • Clinical trial decision support
  • Patient and population decision support
  • Early portfolio decision support
  • Wet lab infrastructure for validation
  • Global network of oncologists and biologists
  • Agentic AI capabilities

Use Cases

  • Accelerating drug discovery
  • Informing clinical trial design and patient selection
  • Generating insights from real-world evidence
  • Automating biopharma R&D processes
  • Understanding disease causality
  • Personalized medicine development

What Physicians Need to Know

Target Identification
Owkin's Discovery AI and TargetMATCH platforms leverage multimodal patient data, including spatial transcriptomics, single-cell and bulk RNA-seq, histopathology images, and whole-exome sequencing, to identify and prioritize novel therapeutic targets. The Discovery AI model has shown superior ability in identifying targets across various cancer types, outperforming DepMap and Open Targets in retrieving known drug targets in clinical phase 2 studies. Owkin's approach focuses on distinguishing true novel targets from known ones, especially in the tumor microenvironment, to reduce attrition in drug development. They utilize a 'Discovery Mode' to surface novel, indication-specific candidates by penalizing dominant signals from cancer hallmarks. Owkin has a proven track record, having delivered multiple AI-discovered candidate drug targets to Sanofi.
Genomic Data Integration
Owkin integrates a wide range of genomic data types, such as whole-exome sequencing, bulk RNA-seq, and single-cell RNA-seq, into its AI models. Their MOSAIC initiative, a large-scale multimodal cancer atlas, incorporates spatial transcriptomics and single-cell analysis from thousands of tumor samples, providing unprecedented insights into tumor and immune cell interactions. This integration allows for a multiscale understanding of disease biology and the identification of multimodal AI targets.
Clinical Trial Matching
Owkin's AI models are designed to match patients to clinical trials based on nuanced eligibility profiles and predict enrollment probabilities. Their K Pro platform helps define optimal patient populations for drug response and identifies responder signatures to inform trial enrichment strategies. Owkin utilizes federated learning to query decentralized patient data across hospital systems, enabling patient identification for trials without moving sensitive records, which is particularly valuable in oncology.
Real-World Evidence Analysis
Owkin emphasizes training its AI models on real-world patient data from a network of over 800 hospitals globally, rather than solely on laboratory datasets. This approach aims to capture the biological complexity, disease heterogeneity, co-morbidities, and treatment variations observed in real patients. They apply machine learning methodologies to optimize clinical development and de-risk clinical trials by leveraging real-world data for prognostic biomarker models and covariate adjustment. Owkin has collaborated with organizations like ASCO's CancerLinQ to analyze real-world oncology data using federated learning to understand treatment resistance.
Literature Mining
Owkin's K Navigator, an AI co-pilot, accelerates literature reviews across millions of scientific articles and draws upon numerous biomedical databases. Through a partnership with Consensus, K Pro users can access research grounded in over 200 million peer-reviewed scientific articles, with controls to refine results based on recency, citations, and journal reputation. This integration helps generate evidence-based insights and supports more precise scientific research workflows.
Collaboration Features
Owkin's federated learning approach allows researchers to collaborate across institutions and leverage diverse datasets without centralizing sensitive patient information, ensuring privacy and data ownership. Their Owkin Studio platform is designed for distributed teams to build, test, and share reproducible machine learning projects, connecting academic datasets to industry use cases under secure governance. Owkin actively partners with leading academic centers, hospitals, and pharmaceutical companies, such as Sanofi, to advance drug discovery and clinical research.
Publication Support
Owkin has a strong track record of publishing its cutting-edge AI research in top-tier journals and at world-leading conferences. Their scientific publications demonstrate their advancements in areas like AI-powered spatial transcriptomics and federated learning for biomedical research.
Physician Tip

Owkin's platforms, particularly K Pro, can significantly accelerate hypothesis generation and validation by providing rapid, evidence-based answers grounded in extensive literature and multimodal patient data. Physicians and researchers can leverage the AI to identify optimal patient populations for clinical trials and understand treatment response at a deeper biological level, including insights from the tumor microenvironment. The use of federated learning ensures patient data privacy while still enabling access to a vast network of real-world data for more robust analyses. The AI-powered digital pathology solutions can streamline diagnoses, screen for predictive biomarkers, and predict patient outcomes, ultimately supporting more informed treatment decision-making.

Owkin's K Pro platform is designed for seamless integration into existing healthcare workflows through an application programming interface (API) service. It is compatible with the Model Context Protocol, allowing researchers to embed AI agents directly into their current systems with minimal disruption. Owkin collaborates with technology leaders like NVIDIA to enhance its biological large reasoning models (OwkinZero) and leverage advanced AI computing ecosystems. They also partner with specialized platforms like Consensus for enhanced scientific literature intelligence. Owkin's solutions are platform-agnostic and plug-and-play, emphasizing scalability and cost savings compared to traditional methods. The company also has SSO integration capabilities for streamlined deployment and user management.

Details

Category Drug Discovery & Research, Oncology AI, Pathology AI
Pricing Unknown
  • The MCP server itself is free to use, but you may need an Owkin account and API access, which may have its own pricing tiers
  • Owkin stock does not trade publicly, but pre-IPO shares might be available through specialized investment platforms
DeploymentOn-premise and cloud-based infrastructure solutions are offered.
API AvailableUnknown
LanguagesEnglish, French
TrainingUnknown
Compliance
BAA AvailableUnknown AI-estimated
HIPAA CompliantUnknown AI-estimated
FDA Status Not applicable AI-estimated

Owkin has CE-IVD approval in the EU for two cancer diagnostics, MSIntuit CRC and RlapsRisk BC. These devices are not FDA cleared or approved in the United States.
GDPRUnknown AI-estimated
Integrations
EHR Not specified
Specialties Genetics, Oncology, Pathology, Pharmacology

Social Proof

Customersunknown
Notable
Sanofi

Support & Reliability

Training ProvidedUnknown

What the Web Says

Owkin is generally viewed as a pioneering company in AI and federated learning for medical research, aiming to accelerate drug discovery and improve patient outcomes. Reviewers highlight its innovative approach to data privacy and collaboration across institutions, though some note the inherent complexities and long development cycles in its field.

Overall: Positive

Strengths

  • Innovative use of federated learning for data privacy and collaboration
  • Potential to accelerate drug discovery and biomarker identification
  • Strong scientific foundation and partnerships with leading institutions
  • Focus on real-world data and clinical applications
  • Addresses critical need for secure data sharing in healthcare
  • Positive impact on medical research and patient care

Limitations

  • Complexity of implementing federated learning in diverse healthcare settings
  • Long development cycles inherent to drug discovery and medical research
  • Reliance on institutional adoption and data sharing agreements
  • Potential for high initial investment in technology and infrastructure
  • Limited public-facing product reviews from individual physicians or small clinics
  • Challenges in demonstrating immediate ROI for all stakeholders

Based on reviews from: Owkin Website, Forbes, TechCrunch, Nature, The Lancet Digital Health, Industry Analyst Reports

Last updated: 2026-09-19

Ratings & Reviews

No reviews yet. Be the first to review this tool!

Rate Owkin

Clinical Value
Ease of Use
Integration
Support & Docs
Value for Money

Press & Coverage

Fierce Biotech
Owkin, Servier ink oncology R&D deal to guide discovery and trial design
AI drug discovery company Owkin has partnered with French drugmaker Servier to license Owkin's AI scientist K Pro for oncology discovery and trial operations. This agreement builds on a 2023 collaboration and allows Servier to access multimodal patient data from Owkin's MOSAIC network for insights into molecule optimization, target discovery, and clinical design.
2026-09
BioSpace
Owkin to License K Pro AI Scientist and Multimodal Oncology and Immunology Data to Boehringer Ingelheim
Owkin announced a license agreement with Boehringer Ingelheim for K Pro, its AI Scientist, and multimodal patient data to accelerate drug discovery in oncology and immunology. This agreement follows a 2025 pilot where Owkin provided spatial insights into the tumor microenvironment.
2026-09
Owkin
Owkin to Build AI Agents as Part of a Multi-Year K Pro Collaboration with Sanofi
Owkin announced a multi-year collaboration with Sanofi to co-develop next-generation biopharma agents, supported by a five-year license for K Pro, Owkin's AI Scientist. This expands on their 2021 strategic partnership focused on target identification in oncology and patient subgrouping.
2026-06
SiliconANGLE
Owkin unveils biology-focused AI agents for clinical research and drug discovery
Owkin has made a series of autonomous AI agents for drug discovery and research available to the healthcare industry, following partnerships with Nvidia Corp. and Anthropic PBC. The company is also releasing its complete agentic infrastructure platform, including a curated multimodal patient dataset from over 800 hospitals.
2026-01
Owkin
Owkin announces Owkin K1.0 Turbigo, a cutting-edge operating system to create the first AGI for biology, at JP Morgan Healthcare Conference
Owkin announced Owkin K1.0 Turbigo, an operating system designed to create the first Artificial General Intelligence (AGI) for biology, at the JP Morgan Healthcare Conference. This initiative aims to transform drug discovery and development.
2025-01
Owkin
Breakthrough in medical research published in Nature Medicine
Nature Medicine published Owkin's research demonstrating the first use of federated learning to train deep learning models on histopathology data from multiple hospitals. This research predicts the response of triple-negative breast cancer patients to chemotherapy while prioritizing patient privacy.
2023-01
PubMed (Eur J Cancer)
An artificial intelligence model predicts the survival of solid tumour patients from imaging and clinical data
This peer-reviewed article details Owkin's development of a prognostic model for solid tumor patients using AI on multimodal imaging and clinical data. The model, built on data from 436 patients and tested on 196, achieved an average concordance index of 0.71 in predicting survival.
2022-08
Owkin
Owkin and Amgen use AI to improve cardiovascular risk prediction
Owkin and Amgen announced the results of a three-year project demonstrating the use of AI to improve cardiovascular risk prediction, published in the European Heart Journal u2013 Digital Health. The machine learning algorithm, trained on data from 13,756 patients, proved more effective than traditional statistical models.
2021-12

Videos

Product demos, reviews, and walkthroughs for Owkin.

View all on YouTube

Frequently Asked Questions

Owkin utilizes a federated learning approach, which means that patient data never leaves the hospital or research institution. Instead of centralizing raw data, Owkin's algorithms travel to the data sources, train on the local data, and then return only the learned model parameters. This distributed training method, combined with techniques like differential privacy and secure enclaves, helps maintain patient privacy and comply with stringent data protection regulations like HIPAA and GDPR.
Owkin leverages a diverse range of real-world data, including electronic health records (EHRs), medical images (histopathology, radiology), genomic data, and molecular data from various research institutions and hospitals. By applying AI and machine learning to this rich, multimodal dataset, Owkin can identify subtle patterns and correlations that may indicate novel drug targets, predict patient response to therapies, and discover new biomarkers for disease diagnosis and prognosis.
Owkin's AI-powered approach offers several advantages over traditional drug discovery, which often relies on hypothesis-driven research and can be time-consuming and costly. By analyzing vast amounts of real-world data, Owkin can generate new hypotheses, identify patient subgroups that may benefit most from specific treatments, and potentially de-risk clinical trials by better predicting drug efficacy and safety. This can lead to a more efficient and accelerated drug development pipeline.
While federated learning offers significant privacy benefits, challenges can arise from data heterogeneity across different healthcare systems, such as variations in data collection methods, coding practices, and patient populations. Owkin addresses these limitations through advanced machine learning techniques designed to harmonize disparate datasets and ensure the generalizability of findings. However, careful validation and interpretation are always crucial when applying models trained on diverse real-world data.
Physicians and research institutions can collaborate with Owkin through various engagement models, often involving participation in research consortia or direct partnerships. Institutions can contribute de-identified real-world data for federated learning projects, gaining access to Owkin's AI tools and insights. These collaborations typically involve formal agreements outlining data governance, intellectual property, and publication rights, ensuring a mutually beneficial research environment.

Related Tools

MyTrials.ai
MyTrials.ai
Drug Discovery & Research
MyTrials.ai is an AI-powered platform for clinical trial recruitment, combining a verified patient network, intelligent matching, and human navigators to accelerate enrollment.
HypeCheck
2 Minute Medicine
Clinical Decision Support & Reference
HypeCheck™ instantly evaluates health articles, wellness advice, and viral trends by cross-referencing them directly against peer-reviewed literature on PubMed, active clinical trials, and official health agency guidance.
$4.99/month or $49.90/year
Scite
Scite
Drug Discovery & Research
Scite is an AI research tool that shows how papers have been cited, providing context on whether findings are supported or contradicted by subsequent research, which is valuable for medical writers and researchers.
Tiered pricing model with free, basic, premium, premium+, group, and enterprise plans.
Google Bard
Google
Clinical Decision Support & Reference
Google Bard, now known as Gemini, is a generative AI model that can assist healthcare professionals with tasks such as summarizing medical records and research papers, generating hypotheses, and providing information on medical conditions.
Freemium
AIDDISON Explorer
Sigma-Aldrich (Merck KGaA)
Drug Discovery & Research
AIDDISSON Explorer is an AI-powered drug discovery platform combining generative AI with computer-aided drug design for molecular generation, virtual screening, ADMET prediction, and retrosynthesis planning.
Noodle Biomedical Literature Discovery
Helena Bioinformatics
Drug Discovery & Research
Noodle is a biomedical literature discovery platform that allows researchers to search publications by meaning, inspect paper details, and explore related evidence through citation connections and semantic neighborhoods. It offers a public read-only Model Context Protocol (MCP) interface for AI agents.
Free

See all Drug Discovery & Research tools →

Suggest an Edit → | Last Verified: 2026-06-11 | First Added: 2026-06-11

Investors who backed Owkin

Funded through the company that built this tool.

Sanofi Ventures
Corporate VC AI T1
Cambridge, United States
$5M-$25M
Sanofi Ventures is the strategic investing arm of one of the world’s largest pharmaceutical companies, and it operates with the patience and…
AI/ML in Healthcare Biotech
Digitalis Ventures
VC T1
New York, United States
Digitalis Ventures is a New York-based healthcare specialist that has built one of the more idiosyncratic franchises in the category by spanning…
AI Tool Finder
AI-powered search. Results may not be comprehensive.