US Public Opinion on AI Policy and Risk (Rethink Priorities Survey)

web

Rethink Priorities·rethinkpriorities.org/research-area/us-public-opinion-of-...

Credibility Rating

4/5

High(4)

High quality. Established institution or organization with editorial oversight and accountability.

Rating inherited from publication venue: Rethink Priorities

This survey is valuable for AI safety advocates and policymakers who need to understand public opinion as a factor in building support for AI governance and risk-mitigation measures.

Metadata

Importance: 52/100organizational reportdataset

Summary

A nationwide poll of 2,444 US adults conducted by Rethink Priorities examining public attitudes toward AI research pauses, regulation, extinction risks, and broader societal impacts. The survey provides empirical data on how the general public perceives AI threats and benefits, revealing nuanced and sometimes surprising views. This data is useful for understanding the political feasibility of AI safety policies.

Key Points

•Surveyed 2,444 US adults on attitudes toward AI regulation, research pauses, and existential/extinction-level risks from AI
•Reveals public perception of AI as both potentially beneficial and potentially threatening to society
•Provides empirical baseline for understanding how AI safety concerns resonate with general (non-expert) populations
•Findings relevant to advocates and policymakers assessing public support for AI governance measures
•Conducted by Rethink Priorities, an EA-aligned research organization focused on evidence-based prioritization

Review

The Rethink Priorities survey provides a comprehensive snapshot of US public sentiment regarding artificial intelligence in April 2023. The study's key contribution is its multi-dimensional exploration of AI perceptions, covering potential risks, regulatory preferences, and expected societal consequences. By employing careful polling methodology and comparing results with previous surveys, the researchers uncovered several noteworthy insights about public attitudes. Methodologically, the survey was strategically designed to replicate and extend previous AI-related polls, using representative sampling and nuanced questioning techniques. Key findings include robust public support for AI research pauses (51%) and regulation (70%), relatively low daily worry about AI's negative effects, and increasing perceived extinction risk over longer time horizons. The research also revealed interesting demographic variations in AI risk perception, with differences emerging across age, gender, and political affiliations.

Resource ID: 6ff0c861116fa036 | Stable ID: Njk4MmEwM2