Anthropic Expands AI for Science With Free Claude Credits for Funders, Institutes and Projects

Anthropic announced on October 28, 2026 that it is expanding its AI for Science program, which gives free Claude credits to researchers working on high-impact scientific projects. The company says it is partnering with 20 scientific foundations, institutions and research groups. The credits go to funders and societies that will pass them on, to research institutions that will use them internally, and to named research projects. The page presents this as a move from supporting individual scientists toward institutional partnerships.

Anthropic started the program in May 2025, saying it believes AI could speed up scientific progress. Over the last few months it also launched a discounted Team plan for scientists. It says it has supported hundreds of projects across biology, medicine, physics, chemistry, climate science and mathematics. Anthropic reports that scientists tell it they work with Claude to generate hypotheses, design experiments, analyze complex data and communicate findings. That is the company's description of what researchers told it, not a measured outcome.

The page sorts the new partners into three groups. Six science funders and societies will distribute credits to researchers: the Alfred P. Sloan Foundation, the Simons Foundation, Schmidt Sciences, the Open Source for Science Fund (a program of Renaissance Philanthropy), the US National Science Foundation and CAS, a division of the American Chemical Society. Six institutions will use credits inside their own organizations: the Space Telescope Science Institute, the Wellcome Sanger Institute, the Francis Crick Institute, the Department of Life Sciences at Imperial College London, EMBL and the Gladstone Institutes. Six research projects receive credits directly.

Two earlier collaborations are used as examples. With HHMI Janelia Research Campus, Anthropic developed the Model Hardware Standard, a shared specification that lets AI agents safely operate lab instruments; it is currently in research preview. Anthropic also supports the Allen Institute, which it describes as building teams of AI agents to analyze large open datasets on the brain, cells and the immune system, turning months of manual analysis into hours. That time claim comes from Anthropic's own page, and the page does not show how it was measured.

The six projects named are FlyWire at Princeton University, which produced the first complete wiring map of an adult fruit fly brain; LEAP at Columbia University, a National Science Foundation center that combines machine learning with physics for climate models; AstroPath at Johns Hopkins University, which images tumors cell by cell using calibration methods from astronomy; gnomAD at the Broad Institute, an open reference of human genetic variation pooled from over 800,000 people; Tau Ceti, a community project building a shared library of mathematics in the Lean proof language; and the Centre for Population Genomics, a joint initiative of Australia's Garvan Institute and the Murdoch Children's Research Institute. Anthropic says many projects will produce public goods such as open datasets, open-source software, open standards and shared libraries.

Several things are not on the page. It does not state the value or volume of credits, any per-grantee limits, or how recipients are chosen by the funders. The introduction says 20 partners. The three lists name 18: six funders and societies, six institutions and six projects. The two earlier collaborators named in the text, HHMI Janelia and the Allen Institute, would bring the named total to 20, but the page does not state how it counts. FlyWire's existing fruit fly brain wiring map is reported, but outcomes attributable to this expanded partnership are not quantified. Anthropic describes many collaborations as early-stage and promises to share their impact later.

For researchers, the practical point is access: the announcement names funders, employer institutions and projects as routes for credits, without saying how these relate to the program’s existing application form. For observers, this is a vendor program that also builds ties with major science bodies. Judging its value will need the follow-up results that Anthropic says it will publish.