A research and advocacy nonprofit that conducts public opinion polling on AI risks and advocates for government policies to mitigate catastrophic risks from frontier AI technology.
A research and advocacy nonprofit that conducts public opinion polling on AI risks and advocates for government policies to mitigate catastrophic risks from frontier AI technology.
People
Updated 05/18/26Executive Director
Funding Details
Updated 05/18/26- Annual Budget
- -
- Current Runway
- -
- Funding Goal
- -
- Funding Raised to Date
- -
Org Details
Updated 05/18/26The AI Policy Institute (AIPI) is a 501(c)(3) research and advocacy nonprofit founded in August 2023 by Daniel Colson, a serial entrepreneur who previously co-founded Reserve (a financial services company) and CampusPA (an executive recruitment firm). Colson launched AIPI after growing increasingly concerned that AI technology was advancing faster than anticipated, particularly following the release of ChatGPT. AIPI's core work centers on three pillars. First, the institute conducts regular public opinion polling through collaborations with leading polling firms like YouGov, surveying US voters to assess concern about AI risks, gauge responses to AI-related news events, and measure support for specific policy proposals. Since its founding, AIPI has conducted more than 20 state, national, and international polls. Second, the institute produces policy research examining threats, regulatory gaps, and potential interventions related to advanced AI, publishing comprehensive reports and policy proposals. Third, AIPI engages with policymakers and media, bridging communication gaps by providing resources on AI technology, public opinion data, and policy options. Within its first few months of operation, AIPI had already met with approximately two dozen lawmakers. AIPI played a notable role in the debate around California's SB 1047, conducting multiple polls demonstrating strong voter support for the AI safety legislation. The institute's polling has been widely cited in media outlets including Axios, Vox, and Politico. AIPI also has a sister organization, the AI Policy Network (AIPN), a 501(c)(4) social welfare organization that conducts direct lobbying for federal AI safety policies. The institute was initially funded by anonymous donors from the tech and finance sectors and has received significant support through the Survival and Flourishing Fund, with grants from Jaan Tallinn totaling $1.935 million across 2024 and 2025 rounds.
Theory of Change
Updated 05/18/26AIPI believes that translating broad public concern about AI risks into concrete, evidence-based policy action is essential to mitigating catastrophic outcomes from frontier AI. By rigorously polling public opinion and demonstrating strong bipartisan voter support for AI regulation, the institute provides policymakers with the political cover and data they need to pursue safety-focused legislation. Through direct engagement with lawmakers, media, and researchers, AIPI aims to ensure that democratic public sentiment is reflected in AI governance rather than allowing the technology's trajectory to be determined solely by industry interests.
Grants Received
Updated 05/18/26Projects
Updated 05/18/26
Discussion
Key risk: Public attitudes toward AI are low-salience and highly framing-dependent, so a polling-led theory of change may not reliably translate into durable, x-risk-reducing federal policy, making AIPI’s counterfactual impact sensitive to execution quality and timing.
Case for funding: AIPI uniquely combines high-quality rapid public opinion polling on frontier-AI risk (e.g., its widely cited SB 1047 surveys and adversarial collaboration work) with direct lawmaker engagement via its c3/c4, giving policymakers credible political cover and ready-made policy options in crisis windows when safety-focused legislation can actually move.