BERI-CHAI Collaboration
About
The BERI-CHAI Collaboration refers to the operational partnership between the Berkeley Existential Risk Initiative (BERI) and the Center for Human-Compatible AI (CHAI) at UC Berkeley. This collaboration has been one of BERI's flagship programs since the organization's founding in 2017. BERI is an independent 501(c)(3) public charity founded by Andrew Critch whose mission is to improve human civilization's long-term prospects for survival and flourishing. BERI's primary activity is collaborating with university research groups working to reduce existential risk by providing them with services and support that are difficult or impossible to obtain through existing university administration channels. CHAI, the Center for Human-Compatible Artificial Intelligence, is a multi-institution research center based at UC Berkeley founded in 2016 by AI pioneer Stuart Russell. CHAI's mission is to develop the conceptual and technical foundations to reorient the general thrust of AI research towards provably beneficial systems. Through this collaboration, BERI has provided CHAI with a wide range of support services including hiring research assistants, machine learning engineers, software developers, technical writers, and administrative staff; purchasing compute resources and facilitating the creation of an in-house compute cluster; paying for conference tickets, travel expenses, software subscriptions, and office equipment; coordinating events and running hiring rounds; and providing general operational support. The collaboration has received substantial funding from Open Philanthropy across multiple grants: $403,890 in July 2017 for core support and CHAI collaboration, $250,000 in 2019 for CHAI ML engineers, $705,000 over two years in November 2019 for continued CHAI collaboration, and $1,126,160 in June 2022 which included facilitating a compute cluster for CHAI and hiring a system administrator. Additional grants supported CHAI internships through BERI. As of 2026, both BERI and CHAI remain active organizations. CHAI continues to conduct AI safety research and hosts annual workshops, with the 10th annual workshop planned for June 2026 at the Asilomar Conference Grounds. BERI continues to provide operational support to multiple university research groups, with CHAI being one of its primary collaborators.
Theory of Change
The BERI-CHAI Collaboration is based on the premise that academic AI safety researchers face significant operational bottlenecks from university bureaucracy that slow down critical research. By providing an independent nonprofit (BERI) to handle hiring, procurement, compute infrastructure, and administrative tasks for CHAI, the collaboration removes these bottlenecks and allows CHAI's researchers to focus on technical AI alignment work. This operational support model enables faster hiring of research engineers and staff, more flexible procurement of compute resources and equipment, and smoother coordination of events and collaborations, all of which accelerate progress on making AI systems provably beneficial and aligned with human values.
Details
- Start Date
- -
- End Date
- -
- Expected Duration
- -
- Funding Raised to Date
- $2,485,050
- Last Updated
- Apr 3, 2026, 1:18 AM UTC
- Created
- Apr 3, 2026, 1:18 AM UTC