Skip to content
Longterm Wiki

Center for AI Safety (CAIS)

Safety Organization
Founded 2022 (4 years old)HQ: San Franciscosafe.ai

Also known as: CAIS

DateEventTypeDescriptionSource
2022"Unsolved Problems in ML Safety" publishedPublicationTaxonomy of open technical challenges in machine learning safety, intended partly as a research agenda for the field.
2022Founded by Dan Hendrycks and Oliver ZhangFoundingNonprofit research organization (EIN 88-1751310) focused on technical AI safety research, field-building, and public communication.
2023MACHIAVELLI benchmark releasedPublicationBenchmark for evaluating goal-directed and deceptive behavior in AI systems.
2023Representation Engineering paper publishedPublicationMethods for reading and steering model internal representations.
May 2023Statement on AI Risk releasedMilestoneOne-sentence statement on AI extinction risk attracted signatures from over 350 AI researchers and industry figures, including Turing Award recipients (Hinton, Bengio, Russell) and CEOs of major AI labs (Altman, Amodei, Hassabis).
2024Reported revenue of $10.2M (FY2024)MilestoneCumulative funding reaches ~$33M since founding ($6.7M in 2022, $16.1M in 2023, $10.2M in 2024).projects.propublica.org (opens in new tab)