The 2025 OECD Survey on Drivers of Trust in Public Institutions, covering OECD member countries and European accession candidates, finds that 35% of respondents expect none of six defined positive outcomes from government AI deployment, making it the single most common response category. Confidence in AI delivering efficiency gains (43% for service improvement, 42% for cost reduction) substantially exceeds confidence in procedural safeguards such as privacy protection (32%), transparency (34%), and fairness (35%), revealing a structural gap between functional and governance expectations. At the individual level, AI optimism correlates with general institutional trust and with self-reported AI knowledge, but high-trust countries do not systematically produce more AI-optimistic publics. For financial institutions assessing government AI procurement and regulatory trajectories, the data suggests that public resistance remains a material friction point for large-scale public-sector AI deployment, with governance and transparency requirements likely to intensify as preconditions for legitimacy.
430 other tracked developments touching the same claims.