Center for AI Safety (CAIS)
- Type
- Civil society organisation
- Place
- United States — United States (San Francisco)
- Last checked
- 2026-09-22
- Next check due
- 2027-03-21
Ways to reach them
- General enquiries
- Media
- Compute cluster access for ML safety researchers — a concrete resource for unaffiliated researchers
This route is retained as reference only. Its verification is missing, more than 180 days old, or the contact window is not currently open; do not rely on it until it is re-checked.
- Careers and fellowships
- Homepage
What it does
Three pillars: technical and conceptual safety research; field-building (fellowships, the ML Safety course, and free access to its compute cluster for safety researchers); and safety standards/advocacy. Authored the one-sentence 2023 Statement on AI Risk signed by most leading AI scientists and lab CEOs. Paired with a separate 501(c)(4), the CAIS Action Fund.
Honest assessment
Well-resourced and reads its inbox. The compute cluster and the ML safety course are the most tangible things any organisation here offers an unaffiliated technical person. Policy input is better routed to the Action Fund.
Concerns it covers
Societal-scale and catastrophic risk — malicious use, AI race dynamics, organisational risks and rogue AI, per its own taxonomy. Covers both misuse and loss of control; broader than MIRI, narrower than PAI.
Government channel
Strong via its Action Fund (see separate entry); CAIS itself was closely involved in California SB 1047 and subsequent state legislation.