The Cloud LAB Act of 2026 directs the National Science Foundation to establish a pilot cloud laboratory network that coordinates NSF‑supported cloud labs with independently operated cloud labs in industry and academia. The statute tasks NSF with designing an implementation plan, standing up an advisory board, and running competitive grant awards to build a distributed infrastructure for remote, automated experiments and standardized biological data generation.
The bill matters because it shifts how experimental capacity is provided: expensive instrumentation and robotic workflows become programmatically available as remotely programmable services, and the federal government will sponsor large, long‑term grants to seed that capability. That creates potential scientific scale‑up and standardized datasets for AI, but also raises difficult questions about data access models, intellectual property, cybersecurity, and minimizing dual‑use risks.
At a Glance
What It Does
The bill requires NSF, in consultation with the Department of Energy and NIST, to run a pilot program that establishes and coordinates a national cloud laboratory network, create an implementation plan, form an advisory board, and award competitive grants to build and operate cloud laboratories. It mandates planning on data storage, publication, access models, IP frameworks, and security considerations.
Who It Affects
Academic labs and research universities that want remote access to advanced wet‑lab automation; private cloud lab operators and biotechnology firms that may seek to join or compete with federally funded nodes; NSF program managers and federal agencies involved in biotech coordination and research security.
Why It Matters
This creates a federal template for treating experimental platforms as shared, remotely accessible infrastructure and for producing curated biological datasets at scale. It alters procurement and funding priorities in biotechnology, sets expectations about public data availability versus subscription/IP models, and requires new governance for biosecurity and cybersecurity across a mixed public‑private network.
More articles like this one.
A weekly email with all the latest developments on this topic.
What This Bill Actually Does
The bill directs NSF to build a network that connects cloud laboratories—with those directly funded by NSF and those run independently—so researchers can find and use capabilities they lack locally. Early work centers on an inventory of existing cloud labs, designing how data will be collected and stored, and building a coordinated approach to standards so experimental outputs from different sites are comparable.
The agency must craft an access model that differentiates nonproprietary, research‑oriented users from commercial subscribers and include sample intellectual property arrangements to govern experimentation and data ownership.
NSF must stand up an advisory board composed of agency staff, academic scientists across biotech subfields, practitioners in biosafety and ethics, and industry representatives. The board's role is to advise on biological data priorities, definitions of authorized researchers, and program design choices that balance usability with security.
The statute requires the board to produce regular guidance and an annual report to inform how the pilot expands.After planning, the program moves into grant rounds to fund multiple cloud lab sites. Awarded sites are intended to operate as nodes that generate high‑quality, machine‑readable biological data for downstream analysis and AI training, and to provide remote experimental access to researchers who otherwise lack instrumentation.
The law also instructs NSF to build security and research‑security considerations into each lab from the outset and to report annually on pilot progress. Finally, the statute is structured as a time‑bounded pilot intended to sunset after a multi‑year period, creating a window for evaluation and course correction.
The Five Things You Need to Know
Within 360 days of enactment NSF must deliver an implementation plan that inventories public and private cloud labs, proposes locations for NSF‑funded labs, and outlines data storage, publication, access, IP, cybersecurity, and cost models.
NSF must establish a cloud laboratory advisory board within 180 days; the board includes federal staff, academic researchers across biotech fields, biosafety and ethics experts, and industry representatives and will terminate 12 years after enactment.
The Director must competitively award grants for at least two Phase II cloud laboratories; each Phase II award must result in a fully operational lab within three years and provide funding for not less than an eight‑year period.
The Director must later competitively award grants for at least three additional Phase III cloud laboratories (separate from Phase II), with awards made no later than four years after enactment and each award providing at least six years of support.
The statute requires the implementation plan to provide an access/payment scheme that allows users doing nonproprietary work to use NSF‑funded cloud labs at no or minimal cost and to make generated data publicly available in a secure, accessible format.
Section-by-Section Breakdown
Every bill we cover gets an analysis of its key sections.
Short title
Names the statute the "Cloud Labs to Advance Biotechnology Act of 2026" or "Cloud LAB Act of 2026." This is administrative but important for citations and program branding; subsequent regulatory materials and solicitations will reference the Act by this title.
Definitions
Provides working definitions for terms the program relies on—artificial intelligence (by reference to an existing statute), authorized researcher, biological data, cloud laboratory, and key officeholders (Director and Under Secretary). Those definitions narrow how the rest of the statute will be interpreted—for example, 'cloud laboratory' explicitly ties the program to physical facilities that combine instrumentation with remote robotic control, which excludes purely in‑silico platforms.
Pilot program and program purposes
Directs the NSF Director, consulting with DOE and NIST, to create a pilot cloud laboratory network that coordinates NSF‑established labs with industry and academic cloud labs. The statute sets dual purposes: to generate high‑quality biological data suitable for AI model training and to provide remote access to instrumentation for research projects. It also charges the program with promoting best practices for data collection, standards, and sharing across nodes, which means NSF will have to develop interoperability and metadata standards as part of the pilot.
Phase I — implementation planning and advisory board
Requires NSF to produce a detailed implementation plan that assesses existing cloud labs, defines payment/subscription and IP frameworks, and includes cybersecurity, biosecurity, and research‑security considerations. NSF must form an advisory board that includes agency and interagency personnel, academic scientists, biosafety and ethics experts, and industry reps; the board advises on data priorities, definitions of authorized researchers, and program expansion. The statute sets a long but finite horizon for this governance body, signaling an expectation of periodic evaluation.
Phase II — competitive awards for cloud laboratories
Authorizes competitive grant awards to establish at least two NSF‑supported cloud laboratories, with each award intended to support multi‑year operations. The law names programmatic milestones and durations for these awards, and requires NSF to ensure awardees reach full operational capability within the startup window. For applicants, this implies readiness to deploy lab automation, data pipelines, and security systems rapidly and to adhere to NSF's access and IP terms.
Phase III — expansion awards
Authorizes a subsequent competitive round to fund additional cloud laboratories beyond Phase II. The Phase III labs are explicitly additive to Phase II and use a similar selection process, allowing NSF to apply lessons learned. The statute prescribes award durations and anticipates adjustments to selection criteria based on the pilot's earlier performance, which gives NSF discretion to elevate operational readiness and security in later rounds.
Reporting and sunset
Mandates annual reports to Congress on pilot progress beginning one year after Phase II awards are all made, and establishes a statutory sunset for the program after a multi‑year period. The reporting requirement creates a formal accountability mechanism and a paper trail of lessons learned; the sunset forces periodic reassessment of whether the network should be continued, scaled, or folded into other agency programs.
This bill is one of many.
Codify tracks hundreds of bills on Science across all five countries.
Explore Science in Codify Search →Who Benefits and Who Bears the Cost
Every bill creates winners and losers. Here's who stands to gain and who bears the cost.
Who Benefits
- Academic researchers and graduate programs — gain remote access to advanced, standardized instrumentation and datasets without buying expensive robotics, enabling experiments and teaching for institutions that lack capital equipment.
- Small biotech startups and early‑stage companies — can lower capital costs by using cloud labs as a service and obtain standardized datasets for model training, accelerating product development and lowering barriers to entry.
- Data scientists and AI developers in biology — receive curated, machine‑readable biological datasets produced under consistent protocols, which improves model training, validation, and reproducibility.
- Federal research ecosystem and interdisciplinary teams — benefit from centralized coordination, standards, and shared resources that can reduce duplication and focus funding on strategic capabilities.
Who Bears the Cost
- National Science Foundation — must absorb program design, oversight, grant management, and long‑term administrative costs associated with coordinating multi‑actor infrastructure and running long‑duration awards.
- U.S. taxpayers/federal budget — funds multi‑year grants and the network’s operational expenses; sustaining the infrastructure beyond the pilot will require continued appropriations or a transition to alternative funding models.
- Private cloud lab operators and academic facilities joining the network — face compliance costs to meet required cybersecurity, biosecurity, data‑format, and access obligations and may have to modify IP or subscription practices to participate.
- University technology transfer and sponsored‑research offices — will need to reconcile sample IP frameworks proposed by NSF with existing institutional licensing practices, potentially complicating commercialization pathways.
Key Issues
The Core Tension
The central dilemma is between maximizing open scientific access and data availability to accelerate discovery (and lower barriers for researchers and educators) versus constraining access and protecting systems to prevent misuse, protect proprietary investment, and manage commercialization—choices that pull the program in opposite directions and that the statute leaves to NSF’s implementation choices.
The bill establishes a clear federal role in creating remotely accessible experimental infrastructure, but implementation raises several knotty trade‑offs. First, standardizing data and opening datasets for AI model training accelerates science but also requires rigorous metadata and quality controls; without strong standards, datasets from different nodes will vary in ways that reduce downstream utility.
Building those standards is complex and resource‑intensive, and the statute leaves substantial technical detail to NSF and its advisory board.
Second, the Act tries to thread a needle on access and IP: it mandates that nonproprietary users have no‑ or low‑cost access to NSF‑funded resources while also permitting subscription models and sample IP agreements. That creates potential conflicts.
Private firms may be reluctant to invest in platform development if they cannot secure exclusivity or favorable licensing; conversely, overly permissive IP rules could let private actors commercialize publicly generated data. The law gives NSF authority to propose frameworks but does not lock in a single approach, which means much will turn on grant terms and implementation guidance.
Third, the program embeds biosecurity and cybersecurity obligations but delegates operational enforcement to awardees and NSF. Effective research‑security screening for 'authorized researchers,' monitoring for dual‑use experiments, and defending distributed lab networks against cyber intrusions are resource‑heavy tasks that require cross‑agency cooperation and possibly new compliance mechanisms.
Finally, the long award durations and statutory sunset imply both commitment and a built‑in evaluation point, but sustaining the network beyond the pilot will depend on future appropriations and whether the pilot demonstrably reduces costs or accelerates innovation.
Try it yourself.
Ask a question in plain English, or pick a topic below. Results in seconds.