NSF PCL-test bed: a cloud lab with AI-integrated remote experimentation (claire)

The National Science Foundation awarded the Trustees of Boston University a cooperative agreement of $4.898 million on August 1, 2026, under the NSF Technology, Innovation, and Partnerships program to develop and operate CLAIRE (Cloud Lab with AI-Integrated Remote Experimentation), an open-access, AI-enabled programmable cloud laboratory for biotechnology.

Boston University will transform its Design, Automation, Manufacturing, and Processes (DAMP) Lab into a remote, cloud-based facility where experiments are performed via automation and natural language interfaces. The system automatically translates simple English instructions into real-world automated experiments by integrating artificial intelligence with biological research. The project encompasses developing an end-to-end bio-foundry pipeline with remote access control policies, biosecurity checkpoints, and standardized community data and protocol exchange.

A core objective is democratizing access to advanced biotechnology capabilities. Boston University will provide subsidized remote access to dozens of new institutions, including community colleges and undergraduate institutions, while maintaining biosecurity standards. The project will distribute open-source protocols and materials to enable similar facilities nationwide, strengthening national biotechnology and AI infrastructure.

Performance of work occurs in Boston, Massachusetts, over a period from August 1, 2026, through July 31, 2030.

Abstract

NSF PCL-test bed: a cloud lab with AI-integrated remote experimentation (claire) this project transforms boston university?s major biological design lab into a cloud lab - an open-source, AI-driven lab in which experiments are performed remotely via automation. By merging artificial intelligence with advanced biology, the project creates a self-driving laboratory capable of automatically translating simple english instructions into real-world automated experiments. This initiative represents a vital national investment because IT directly enhances national security and economic resilience by accelerating discoveries in many fields, such as enabling local manufacturing of technologies using rare earth elements and lowering the cost of developing life-saving therapeutics. Furthermore, IT advances the national science foundation's mission to foster innovation and train the future american workforce. A cornerstone of this project is its commitment to democratizing science; the lab will offer subsidized, remote access to dozens of new institutions, including community colleges and undergraduate institutions, while maintaining the highest biosecurity standards. Through the distribution of materials made using open-source protocols, this project will allow similar facilities to be developed nationwide to strengthen our national biotechnology and AI infrastructure. The claire project establishes an open-access, AI-enabled programmable cloud lab node for biotechnology by transforming the remote, modular, and composable BIO-cloud operations of boston university's design, automation, manufacturing, and processes (damp) lab. The project focuses on developing and deploying an end-to-end BIO-foundry pipeline that integrates remote access control policies, biosecurity checkpoints, and standardized community data and protocol exchange. To drive this academic core facility, natural language-based specifications and AI-agent composable extensions will be applied to three key domains: protein function discovery, critical mineral asset discovery, and therapeutic biologics. Specifically, our methods utilize growth-based quantitative sequencing for massive-scale protein variant data collection; engineering microbial biosensors based on the lanmodulin protein for rare earth element detection in?pseudomonas putida; and applying mechanistic and?in-silico?models to predict therapeutic protein expression and developability. To power this automation, human-readable experimental descriptions will be translated into standardized executable code via a domain-adapted large language model (LLM), continuously improving through reinforcement learning techniques. In addition, resource use will be optimized through automated job scheduling. The intellectual merit of this work lies at the convergence of AI, self-driving labs, and synthetic biology to democratize access to replicable, secure, remote biotechnology r & d. Ultimately, the project will deliver open-source software, automated protocols, and trained AI agents to establish reproducible, modular, and autonomous workflows, generating high-quality public datasets to train future predictive models in protein engineering and biomanufacturing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are planned for this award.