
AIDO
Multiscale foundation models for predicting, simulating, and programming biology across molecules to phenotypes.
Overview
AIDO (AI-Driven Digital Organism) is a groundbreaking world model for biology developed by GenBio AI, designed to connect and simulate biological information across scales — from molecules to phenotypes and ultimately to whole individuals. Built on a system of multiscale AI foundation models, AIDO aims to make biology computable, predictable, and programmable, enabling researchers, clinicians, and biotech innovators to accelerate breakthroughs in medicine, drug discovery, and life sciences.
GenBio AI was co-founded by Eric Xing and David Baker, with a shared vision that simulating biology with precision will transform disease treatment from a decades-long gamble into an engineering problem. The platform is intended for researchers, pharmaceutical developers, and life sciences organizations seeking to leverage cutting-edge generative biology to advance their work.
Core Capabilities
- Multiscale biological modeling that spans from molecular-level interactions to organism-level phenotypes
- Prediction, generation, and simulation of biological processes, including virtual cell behavior
- A programmable, interactive digital living system that models the full complexity of human biology
- State-of-the-art performance across more than 300 biological tasks
Foundation Models
- K2 — The most advanced foundation model, with 70B+ parameters
- World's Largest Genome Foundation Model — 15B+ parameters, released in 2024
- World's Largest Protein Foundation Model — 100B+ parameters, released in 2024
- World's Largest RNA Foundation Model — 7B+ parameters, released in 2024
Key Use Cases
- Drug Discovery: Accelerate drug development by simulating millions of potential treatments and predicting their effects on complex biological systems
- Disease Understanding: Equip researchers with models to investigate the mechanisms underlying conditions such as cancer and neurodegenerative diseases
- Personalized Medicine: Enable digital patient simulations to minimize adverse reactions and improve the effectiveness of individualized treatments
Research and Publications
- Foundational paper: Toward AI-Driven Digital Organism: A System of Multiscale Foundation Models for Predicting, Simulating and Programming Biology at All Levels
- Research on accurate and general DNA representations emerging from genome foundation models at scale
- Work on Mixture of Experts approaches enabling efficient and effective protein understanding and design
GenBio AI's models are available for download via Hugging Face and GitHub, making them accessible to the broader research community. Beyond healthcare, the AIDO Digital Living System is positioned to deliver actionable insights across green technologies and consumer innovation, reflecting the platform's broad applicability across life sciences and beyond.

