🔍 Read the full analysis: 5 Ways Foundations And Institutions Are Engaging With AI For Science on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Anthropic says it is expanding its AI for Science program to include 20 scientific foundations, institutions and research groups. The announcement names 12 partners and describes Claude credits and possible shared resources, but does not specify funding amounts, project outcomes or the other eight organizations.
As described in the original analysis, Anthropic says 20 scientific foundations, institutions and research groups are joining its AI for Science program, expanding the company’s effort to put Claude credits in the hands of research organizations. The company says participating funders will distribute credits to researchers and institutional partners will use them internally, but it has not disclosed the value of the credits or specific project results.
Anthropic named six funders and scientific organizations that will distribute credits: the Alfred P. Sloan Foundation, the Simons Foundation, Schmidt Sciences, the Open Source for Science Fund at Renaissance Philanthropy, the U.S. National Science Foundation and CAS, a division of the American Chemical Society. The announcement also identifies six institutions that will use credits within their organizations: the Space Telescope Science Institute, Wellcome Sanger Institute, Francis Crick Institute, Imperial College London’s Department of Life Sciences, the European Molecular Biology Laboratory and Gladstone Institutes.
Although Anthropic describes a total of 20 organizations, the supplied announcement material names 12 and does not identify the other eight. It says the program may also support resources for wider use, including open datasets, software, standards and libraries. Those are potential outputs, not confirmed commitments tied to particular partners.
Anthropic says AI for Science began in May 2025 to increase researchers’ access to AI tools and resources. The company reports supporting hundreds of projects across biology, medicine, physics, chemistry, climate science and mathematics. It has not provided a project breakdown or an independent evaluation of the work. Anthropic has also introduced a discounted Team plan for scientists, according to the source material.
Anthropic expands research access through institutions
Anthropic says 20 organizations will distribute or use Claude credits for scientific research. Credit amounts and project outcomes have not been disclosed.
How support reaches researchers
Partners will distribute credits or use them within their organizations.
- Distribute credits to researchers
- Six organizations are named
- Use credits within their organizations
- Six institutions are named
Research spans many fields
Anthropic says it has supported hundreds of projects, without providing a project breakdown or independent outcome evaluation.
Potential shared resources
Some projects may create materials that other researchers can reuse.
Partners allocate credits to research teams or use them in-house.
Projects may produce open datasets, software, standards or shared libraries.
The announcement does not specify committed outputs, release timelines or evaluation methods.
What remains unclear
The announcement leaves key details open for researchers and readers.
- Credit amounts and support duration
- Allocation dates and project milestones
- Names of all 20 organizations
- Independent evidence of research outcomes
- Whether resulting resources will be openly available
Research Access Moves to Institutions
The expansion shifts the program beyond support for individual researchers toward organizations that fund, host or coordinate research. Those partners may be positioned to connect AI tools with established teams and research infrastructure. Whether that access improves research speed or quality remains untested in the information Anthropic provided.
Potential shared outputs could matter beyond participating groups. If partners release datasets, software or standards for broader use, other researchers may be able to build on them. But the announcement does not name specific deliverables, release dates or access terms, so the scale of any public benefit is not yet known.
The announcement also leaves open practical questions about resource allocation: how many researchers will receive credits, how much support each organization gets, and how long it lasts. Those details will shape how meaningful the expansion is for research teams with different needs and resources.
Products Worth Considering
Apple MacBook Air 13-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Liquid Retina Display, and Touch ID – Midnight
Sonos Era 100 Wireless Smart Speaker, Black
Apple iPad 11-inch (A16 Chip, 128GB, Wi-Fi, Pink)
As an affiliate, we earn on qualifying purchases.
Earlier AI for Science Work
Anthropic says participating scientists have used Claude to generate hypotheses, design experiments, analyze complex data and communicate findings. These are company descriptions of reported uses; the material does not include independent evidence assessing effects on scientific results.
The announcement points to earlier collaborations as examples of institutional work. With HHMI Janelia Research Campus, Anthropic has worked on a Model Hardware Standard, a shared specification intended to let AI agents operate laboratory instruments safely. Anthropic describes the standard as being in research preview. The company also says the Allen Institute is building teams of AI agents to analyze research data, with the goal of reducing tasks that can take months to hours. The material supplies no independent findings or detailed evaluation of either effort.
“Our goal is to increase access to AI tools and resources to enable researchers to deepen and expand their own scientific work.”
— Anthropic
Credit Amounts and Results Unknown
The announcement does not disclose credit values, allocation schedules, support duration or the number of researchers expected to benefit. It names 12 partners despite describing a 20-organization expansion, and the supplied material does not list the other eight or complete its introduction of additional research initiatives.
It is also unclear which partners will produce public datasets, software or standards, and how any resulting resources will be managed. Anthropic provides no evaluation methods or reporting requirements for partner projects, and no independent evidence here measures scientific quality, reproducibility or time saved.
Partners Begin Credit Allocations
Anthropic says funders and societies will distribute Claude credits, while institutional partners will use credits within their organizations. The company has not announced allocation dates, project milestones or a schedule for reporting results.
Further details from Anthropic and its partners could clarify which studies receive support and whether projects release reusable resources. Evidence on research methods, findings, reproducibility and access to resulting tools or datasets will help readers judge the program’s effects beyond the participating organizations.
Key Questions
What did Anthropic announce?
Anthropic said its AI for Science program is expanding to 20 organizations, which will distribute or use Claude credits for scientific work.
Which organizations are named?
The supplied material names 12 partners: six funders or scientific organizations that will distribute credits and six research institutions that will use them internally. It does not identify the other eight organizations in the announced total.
How much funding or credit will each partner receive?
No amounts are disclosed. Anthropic has not specified credit values, allocation dates, support duration or how many researchers will receive access.
Has the program shown that Claude improves scientific results?
The announcement describes reported uses and projects, but provides no independent evaluation of scientific outcomes, research quality, reproducibility or time saved.
Will the work produce resources for the public?
Anthropic says projects may produce open datasets, software, standards or libraries. These are possible outputs; the announcement does not identify confirmed deliverables, release schedules or access terms.
Primary source: Anthropic · via ThorstenMeyerAI.com
