Field builder · Research orgs · Berkeley, California · Est. 2023
CivAI (Civic AI Security Program) is a California-based, independent 501(c)(3) nonprofit focused on “building public understanding of AI” through live, interactive software demonstrations rather than relying primarily on research papers or abstract discussion. Its public-facing tools are designed to give users a direct, “gut-level” experience of AI capabilities and risks—particularly in areas such as cybersecurity (e.g., phishing), deepfakes, elections, and other frontier AI safety concerns—so that policymakers and civil-society audiences can better evaluate real-world implications.
CivAI also develops additional briefing formats for leadership audiences and, for higher-risk cybersecurity content, provides demos to “approved organizations” via request. Beyond demos, the organization participates in policy and standards efforts, including membership in the Artificial Intelligence Safety Institute Consortium (AISIC) established at NIST, where it has described work translating NIST AI Risk Management Framework guidance into practical steps for state and local governments.
Publications are not yet indexed for this organization. Read them on civai.org ↗.