Organisation
Center for AI Safety
A nonprofit combining technical AI-safety research, benchmarks and field-building with public advocacy; author of the 2023 Statement on AI Risk.
The Center for AI Safety (CAIS) is a San Francisco nonprofit, founded in 2022 by Dan Hendrycks, that combines technical AI-safety research with field-building and public advocacy. It drew wide attention in May 2023 with a one-sentence Statement on AI Risk — that mitigating "the risk of extinction from AI" should be a global priority alongside pandemics and nuclear war — signed by hundreds of researchers and executives including the heads of OpenAI and Google DeepMind, which helped move existential-risk arguments into mainstream political debate. CAIS also produces benchmarks and research on model safety, runs grant and compute programmes for the field, and its director Dan Hendrycks has advised companies including xAI and Scale AI.
- Category
- Safety & alignment research
- Founded
- 2022
- HQ
- San Francisco, US
- Key people
- Dan Hendrycks
Appears alongside
Featured in threads
Tracks
- Benchmarks & progress 2
- Safety & alignment 2
- Ideas & essays 1
Center for AI Safety releases MASK honesty benchmark
Built with Scale AI, the benchmark found models that scored well on truthfulness tests still lied readily under pressure, and that larger models did not become more honest.
Benchmarks & progress · Safety & alignment
CAIS and Scale AI unveil Humanity's Last Exam results
A 2,500-question expert benchmark built from submissions by nearly 1,000 academics found every frontier model, including o1 and GPT-4o, scored under 10%.
Benchmarks & progress
Lab leaders sign a one-sentence statement on extinction risk
"Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war."
Ideas & essays · Safety & alignment
Also mentioned in 1 entry
Referenced in passing — Center for AI Safety isn't the main subject of these.