AI Science Foundry
Advancing the Science and Engineering of the Future
Carnegie Mellon University’s AI Science Foundry harnesses the power of autonomous, AI-enabled laboratories to accelerate the innovation-to-impact timeline from years to months.
CMU’s Transformative Programmable Cloud Lab
The AI Science Foundry brings together AI, automation, robotics and computational workflows with large-scale computing, data storage and access to experimental programmable cloud lab facilities in chemistry, biology and hard materials. These work in concert, allowing researchers and engineers from across the nation to remotely design and deploy autonomous science experiments to collect and analyze data.
Driving Innovation Together
Developing this robust platform, and integrating it in a national network, will require universities, national labs and industry to collaborate in building unique infrastructure, tools, software standards and instruments. Achieving this vision demands the ability to execute and adapt complex workflows quickly and reliably as new ideas emerge. No single organization can accomplish this alone. Collaboration across institutions, disciplines and technologies is essential to build the infrastructure that will drive the next era of discovery.
What is a Programmable Cloud Lab?
A programmable cloud lab is an AI-guided autonomous experimentation lab where scientists remotely design, control, and execute complex experiments using robotics, making high-tech science more accessible, efficient, and collaborative.
Retaining a U.S. Competitive Edge
Autonomous experimental labs are being deployed in industrial, academic and national lab environments, but most of these remain highly dedicated to specific scientific outcomes and have limited use of AI. Other countries are developing AI-powered laboratories and automated scientific infrastructure at a national scale. For the US to retain our competitive advantage, we must harness the power of AI – through automated laboratories like the AI Science Foundry - to drive scientific innovation at a pace faster than at any time in human history.
One-of-a-Kind Platform
The Accelerating Scientific Autonomous Platform (ASAP) combines software-controlled instruments, multi-purpose autonomous robots, and sensors to run multiple AI-guided programmable workflows in parallel.
Three Science Drivers
The AI Science Foundry enables AI-guided autonomous experimentation using the nation’s largest collection of co-located research tools for hard materials, chemistry, and biology across two facilities.
Transforming Science
Researchers using the AI Science Foundry have access to sophisticated instrumentation that enable them to conduct scientific inquiries that previously exceeded their capabilities. AI agents design and orchestrate experiments, drawing on any discipline or any multi-disciplinary domain by consulting published literature, health records, images, and the Foundry’s own expanding knowledge base. AI agents optimize experiments and comb vast repositories of data that no human can totally comprehend. As they analyze data, they perceive correlations and causality that scientists might not recognize.
Carnegie Mellon: Where the Future of Science Begins
Carnegie Mellon’s expertise in AI, robotics, and automation—combined with a legacy of innovation and interdisciplinary collaboration— have long helped accelerate innovation and make ambitious, cross-cutting research possible. The AI Science Foundry leverages applied AI in software, hardware and cybersecurity resources at CMU’s National Robotics Engineering Center and the Software Engineering Institute. It builds on CMU’s first-of-its-kind, state-of-the-art facilities, serving as both immediate research platforms and ongoing test beds shape the future of scientific progress.
Contact us:
Reach out to our team to learn more the Carnegie Mellon AI Science Foundry: AIScienceFoundry@cmu.edu