From the SLED-wide archive edition of August 29, 2026
Cross-national survey finds deep skepticism toward government AI
Organisation for Economic Co-operation and Development · Public administration and citizen trust · OECD member and European accession-candidate countries
- Publisher
- Trustworthy artificial intelligence in the public sector
- Original publication
- June 29, 2026
- Source retrieved
- Not recorded in the historical archive
What happened
The OECD's 2025 trust survey found that 35% of respondents across participating OECD countries expected none of six positive outcomes from government AI use, while only 22% held very positive expectations.
Why it matters
State and local AI services operate in a trust environment that can determine adoption, resistance, legal challenge, and perceived legitimacy even when a system is technically accurate.
Evidence and measured results
Skepticism rose from 24% among people aged 18–29 to 41% among those 50 and older and was higher among people with lower formal education, financial insecurity, or perceived discrimination. Only about 32% expected government to protect personal information from unauthorized access or misuse when using AI, twenty points below confidence in legitimate government data use generally.
Limitations and uncertainty
The survey measures expectations and perceptions, not observed system performance or causal effects. Country averages conceal large national and local differences, and attitudes may change with direct experience.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source as summarized in the preserved archive. Enriched 2026-09-05; this does not change the original publication date. Labels below come from the analysis itself.
Sales
Role takeaway
- Customer problem
- a technically functioning public AI service may face skepticism about privacy, accountability, and fairness.
- Stakeholders
- service leadership, community engagement, privacy, accessibility, communications, and groups affected by the proposed use.
- Discovery
- can residents understand the purpose and data flow; what human appeal or alternative exists; and whose concerns are missing from current engagement?
- Value hypothesis
- transparent design and meaningful recourse may support a more trustworthy experience, to be tested locally.
- Potential engagement
- stakeholder discovery and a service-transparency/usability assessment.
- Unsupported claims
- the survey's expectations are not performance results, demographic associations are not causal explanations, and country averages cannot predict local trust or prove that any proposed intervention will improve it.
Pre-sales engineering
Role takeaway
- Fit
- apply the findings to the user experience and data handling of a specific public AI service; the survey does not establish a technical product preference.
- Architecture and integration
- expose purpose, data use, human ownership, appeal routes, and accessible alternatives within existing service channels.
- Prerequisites
- accurate data-flow documentation, enforceable purpose limits, and responsible privacy/community-engagement owners.
- Constraints
- aggregate trust scores can conceal group differences, and collecting demographic feedback creates its own privacy responsibilities.
- Security
- minimize data, verify access controls and breach readiness, and ensure public explanations match actual processing.
- Proposed validation
- test comprehension, usability, and recourse with affected groups, measure attitudes separately from system accuracy, and avoid unnecessary sensitive feedback collection.
Delivery
Role takeaway
- Work
- engage affected communities, document the service's purpose and data handling, implement notice and recourse, and review trust alongside operational results.
- Dependencies
- accessible channels, service-owner participation, valid privacy explanations, and capacity to respond to appeals.
- Ownership
- the service leader owns legitimacy and accountability; privacy/security teams validate handling; community and accessibility leads design inclusive engagement.
- Skills and adoption
- train frontline staff to explain the AI role and offer meaningful alternatives rather than pressuring use.
- Governance checkpoints
- predeployment engagement, impact review, and periodic feedback assessment.
- Proposed acceptance
- users can identify purpose and human recourse, alternative paths work in tests, and feedback is reviewed across relevant groups using privacy-approved methods. Risks include token consultation and equating adoption with trust.
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Expose purpose, data use, human accountability, appeal routes, and service alternatives in the experience; collect trust and usability signals by demographic group; and make privacy-preserving design visible rather than purely contractual.
Governance
Who approves, reviews and stays accountable for outcomes?
Treat public legitimacy as an outcome measure. Engage affected communities before deployment, publish impact and evaluation evidence, provide meaningful notice and recourse, and avoid using adoption rates as a proxy for trust.
Security and privacy
What data, permissions and controls need testing?
The gap between general confidence in government data use and confidence under AI makes data minimization, access control, breach readiness, explainable data flows, and enforceable purpose limitation central to adoption.
The preserved archive analysis covered architecture, governance and security. Not assessed for this record: accessibility and workforce, procurement, operating model.
Publication history
- 2026-08-29SLED-wide archive · Issue 0214 resources
Stable resource ID: oecd-public-trust-ai