Anthropic Launches $5M Wellbeing Grants and Opens Claude Data to Researchers
Research into AI's effects on users has until now been confined to AI labs, which cannot independently audit their own products. These programs extend that access to outside researchers.
Anthropic launched a $5 million grant program for independent research on how AI affects users' wellbeing, and gave external researchers access to real, privacy-preserved Claude usage data for the first time. The grant program provides direct funding, model access, and technical support; grantees work independently and publish open-source evaluations any developer can use. The interest deadline for the grant program is September 14. Anthropic also ran a pilot of the data access program, published its results, and is deliberating how to scale it without compromising user privacy.
Anthropic on September 1 announced free Claude Team subscriptions for verified principal investigators, offering Standard seats at no cost and Premium seats at $15 per month for research teams for up to one year, covering 10,000 scientists worldwide.
The program is a third research support initiative alongside the $5 million wellbeing grant program, whose interest deadline is September 14, and the external data access pilot.
Anthropic on August 28 published results from a completed pilot giving external researchers access to real, privacy-preserved Claude usage data, and said it is deliberating whether the program can be scaled without compromising user privacy.
The pilot accompanies a $5 million grant program for independent research into how AI affects users' wellbeing; grantees receive direct funding, Claude model access, and technical support, and publish their work as open-source projects.
Anthropic on August 28 announced 10,000 Claude Team seats for scientists, with standard seats free and premium seats at $15 per month, an 80% discount, for one year.
The AI for Science credit program is also expanding beyond biological sciences to offer up to $50,000 per project; researchers in biology and chemistry are currently limited to Opus-class models, and Anthropic said it is working with the US government on Mythos-class model access for life sciences. The grant program's interest deadline is September 14.
Anthropic announced on August 26 a $5 million grant program for independent research into how AI affects user wellbeing, and for the first time gave external researchers access to real, privacy-preserved Claude usage data.
The grants include direct funding, model access, and technical support; grantees work independently and publish results as open-source projects any developer can use. Anthropic said "to date, this work has only been possible within AI labs" and that "we can't tell the whole story alone, so we opened up our tools".
I send one email each morning with the stories that moved. If you would rather just read here, that works too.
Subscribe free