AI for Science expands to foundations and institutions

Why it matters

Anthropic is expanding its AI for Science program from individual researchers to institutional partnerships, offering free Claude credits to a globally distributed set of scientific foundations, institutes, and research groups. Launched in May of last year, the program aims to broaden access to AI tools for high-impact research across biology, medicine, physics, climate science, and other fields. Two partnerships illustrate the shift toward infrastructure and collaboration. With HHMI Janelia Research Campus, Anthropic developed the Model Hardware Standard, a shared specification that lets AI agents safely operate laboratory instruments, now in research preview. The Allen Institute, known for open datasets on the brain, cells, and the immune system, is building agent teams to analyze that data, reportedly shrinking months of manual work to hours. Partners include science funders like the Simons Foundation, Schmidt Sciences, and the US National Science Foundation, plus astronomy and genomics institutes such as the Space Telescope Science Institute and the Wellcome Sanger Institute. Many projects are expected to yield public goods like open datasets and shared standards.