Stanford Institute for Human-Centered Artificial Intelligence (HAI) — program: Center for Research on Foundation Models (CRFM) — studies foundation model capabilities, risks, and societal implications. Approximately 175 affiliated researchers.

NoteThe source confirms the core claim: CRFM is a program of Stanford HAI, it studies foundation models and their capabilities/risks/societal implications (explicitly mentioned: 'Technical foundations,' 'Beneficial applications,' 'Societal impact,' 'AI policy'). However, the source does not mention the specific figure of '175 affiliated researchers as of 2024.' The source only states '10+ departments' are represented, which is not a researcher count. The HELM benchmark is also not mentioned in the provided excerpt. Since the researcher count cannot be verified from the source text provided, the claim is only partially confirmed.