OpenAI and Anthropic back a $15 million Simons math effort, and Rosalind opens to US labs

On October 15, the Simons Foundation said it, Anthropic, and OpenAI will commit $15 million in combined funding and resources. The foundation will manage and award it, including through an open call, for workshops and other work that helps mathematicians check AI-assisted results, learn the tools, and set standards for using them. Anthropic and OpenAI will also provide tools and computing to researchers the foundation supports, including outside mathematics.

OpenAI says it is contributing $5 million, plus model and compute access for foundation-supported researchers. Anthropic says it has committed $5 million of funding, compute, and Claude access. The foundation page states the combined $15 million. It does not print the split. The $5 million figures are the companies' own posts.

Separately, OpenAI opened the Rosalind Academic Plan. Eligible labs at US universities can receive up to 30 ChatGPT Business Standard seats at no cost for up to 12 months. GPT-Rosalind access still needs trusted-access approval. Applying does not turn access on. Labs at a university that already has an OpenAI Enterprise agreement use that university's process. The help page says the offer does not include API access or API credits. The release notes carry the same terms.

What it costs

The Simons money is a grant program, not a discount on ChatGPT or Claude. The Rosalind seats are $0 for up to 30 people, for up to a year, if the lab is approved. Everyone else pays the normal Business Standard price.

Should you switch

Mathematicians: the open call is the Simons piece to watch, and it is not a form you can submit from these posts yet. The foundation said it will award the funding, including through an open call. US university labs that want GPT-Rosalind apply through the academic plan, from a principal investigator or lab lead's institutional account. Use is still under OpenAI's Usage Policies.