Representative End User Clients
Representative Automation Clients
Representative Software Clients
The new application is designed to help life sciences researchers accelerate drug discovery and antibody development using AI-driven workflows.
At its AWS Life Sciences Symposium 2026, held in New York City with an attendence of over 2,500, Amazon Web Services announced Amazon Bio Discovery, a new AI-powered application designed to help scientists design and test drug candidates more efficiently during the early stages of drug discovery.
The platform provides researchers with access to biological foundation models (bioFMs), specialized AI models trained on biological datasets to generate and evaluate potential drug molecules. The application is intended to support antibody therapy development by helping scientists identify and refine candidate molecules more quickly.
AI Agent Supports Drug Discovery Workflow
Amazon Bio Discovery includes an AI agent that enables researchers to interact with the platform using natural language to:
Select models for research goals
Optimize experimental inputs
Evaluate candidates for experimentation
Design experiment workflows
Scientists can also train models on prior experimental data to improve prediction accuracy and benchmark models using antibody datasets to assess manufacturability, stability, and biological suitability.
Addressing AI Adoption Challenges in Research
AWS designed the platform to help address several barriers currently limiting AI adoption in drug discovery, including:
Difficulty selecting and benchmarking AI models
Limited access to computational biology expertise
Infrastructure complexity
Manual workflow coordination across disconnected systems
Amazon Bio Discovery is designed to help simplify these processes through benchmarked AI models, analysis packages, and integrated workflow tools.
Integrated Lab Partner Network
To support end-to-end experimentation, Amazon Bio Discovery helps to allow scientists to send selected candidates directly to integrated laboratory partners for synthesis and testing.
Current integrated partners include:
Twist Bioscience
Ginkgo Bioworks
A-Alpha Bio is expected to join the platform in the future.
Test results are routed back into the application to help support iterative refinement and create a lab-in-the-loop experimentation cycle.
Custom Model Training and Internal AI Support
Amazon Bio Discovery helps to enable organizations to fine-tune AI models using proprietary experimental data. Researchers can securely upload prior lab data and train custom models without building dedicated machine learning pipelines.
For organizations with existing internal AI capabilities, the platform also helps to support deployment of in-house models within the Amazon Bio Discovery environment.
Early Adoption and Testing Results
Memorial Sloan Kettering Cancer Center used Amazon Bio Discovery during testing to design approximately 300,000 novel antibody molecules, helping to reduce what would traditionally take months or longer into a process that was completed within weeks.
Early adopters of the platform also include:
Bayer
Broad Institute
Fred Hutch Cancer Center
Voyager Therapeutics
Follow ARC Advisory Group for the latest trends in sustainability, automation, and next generation technology.