Organization

Center for AI Safety (CAIS)

Nonprofit research · Research lab · San Francisco, California

The Center for AI Safety is a San Francisco-based research and field-building nonprofit whose stated mission is to reduce societal-scale risks from artificial intelligence. Its work covers technical safety research, including benchmarks, as well as field-building and advising industry and policymakers.1 2 4

Last reviewedSep 29, 2026Entry updated Sep 29, 2026

Key facts

Website: safe.ai (external site: safe.ai)
Legal name: Center for Artificial Intelligence Safety Inc3 8
Ownership: Nonprofit
Legal form: Corporation recognized by the IRS as tax-exempt under Section 501(c)(3); the IRS extract classifies it as an organization receiving substantial support from a governmental unit or the general public (170(b)(1)(A)(vi)).3 8 9
Headquarters: San Francisco, California1 8
Founded: Unknown
Sectors: Research
Careers: Careers page (external site: safe.ai)

What is and isn’t public

An evidence-based summary of what this catalog has documented. It is not a verdict on the organization as a whole.

Publishes research papers and benchmarks. Humanity's Last Exam, for example, has MIT-licensed evaluation code on GitHub and a question set on Hugging Face that can be downloaded after an automatic access prompt, with a private held-out set kept back. See the Humanity's Last Exam record.5 6 7 4

Products

Documented products and how they are delivered. Each row links to the official page.

No product records yet.

Related open artifacts

Records in the Open Models & Tools directory that list this organization as a maintainer or publisher.

Evaluation tools (1)

Humanity's Last Exam (HLE)Evaluation toolAvailability: Partial

Notable documented facts

CAIS's research page lists benchmarks it has published, including Humanity's Last Exam, the WMDP benchmark for hazardous knowledge, and HarmBench for automated red teaming.4
Humanity's Last Exam was organized by a team from CAIS and Scale AI, with questions contributed by subject experts at institutions in many countries.5

U.S. eligibility

How this record meets the catalog’s published eligibility policy.

Eligible · basis: U.S. nonprofit or lab

CAIS describes itself as a San Francisco-based nonprofit, and its donation page states that the Center for Artificial Intelligence Safety is a U.S. 501(c)(3) organization with EIN 88-1751310. The IRS exempt-organization extract for California lists that EIN as Center for Artificial Intelligence Safety Inc at a San Francisco address, with 501(c)(3) status.1 3 8

Assessed Sep 29, 2026

Eligibility policy

Sources

1.
Center for AI Safety (CAIS) (external site: safe.ai)
Center for AI Safety · Official page · accessed Sep 29, 2026
2.
About | CAIS (external site: safe.ai)
Center for AI Safety · Official page · accessed Sep 29, 2026
3.
Donate | CAIS (external site: safe.ai)
Center for AI Safety · Official page · accessed Sep 29, 2026
4.
Research | CAIS (external site: safe.ai)
Center for AI Safety · Official page · accessed Sep 29, 2026
5.
Humanity's Last Exam (external site: lastexam.ai)
Center for AI Safety and Scale AI · Official page · accessed Sep 29, 2026
6.
centerforaisafety/hle LICENSE (external site: raw.githubusercontent.com)
Center for AI Safety · License · accessed Sep 29, 2026
7.
cais/hle (Hugging Face dataset API metadata) (external site: huggingface.co)
Hugging Face · Other · accessed Sep 29, 2026
8.
Exempt Organizations Business Master File Extract - California (eo_ca.csv) (external site: irs.gov)
Internal Revenue Service · Filing · accessed Sep 29, 2026
9.
Exempt Organization Business Master File Extract information sheet (external site: irs.gov)
Internal Revenue Service · Documentation · accessed Sep 29, 2026

Report a correction Edit this entry

This listing is not an endorsement or a federal approval.