{"resourceId":"new-america-ai-government-field-review","versions":[{"version":"legacy/2026-09-01/new-america-ai-government-field-review","resource":{"id":"new-america-ai-government-field-review","title":"Field review finds SLED experimentation broadening faster than capacity, strategy, and outcome evidence","organization":"New America","sector":"State and local government","geography":"United States","publishedAt":"August 31, 2026","sourceName":"AI in Government: A Field-Level Review","sourceLabel":"New America field review","sourceUrl":"https://www.newamerica.org/insights/making-ai-work-for-the-public/ai-in-government-a-field-level-review/","evidenceClass":"independent-research","outcomeClass":"mixed","topics":["knowledge-work","developers-agents","infrastructure","governance-procurement","accessibility-workforce","operating-model"],"finding":"New America's review combined more than 40 practitioner and expert interviews, pilot work, literature review, legislation analysis, and a field scan of state and city activity. It found rapidly expanding legislative and pilot activity, with states favoring enterprise sandboxes or walled gardens and cities favoring stand-alone service pilots, but local capacity, coherent strategy, trust, infrastructure, budgeting, and evidence of return remained major constraints.","sledRelevance":"This is a useful operating-model snapshot of why visible experimentation does not automatically become durable public capability. It also highlights the distinct paths of states and cities and the potential role of universities, professional associations, and shared institutions in filling capacity gaps.","evidence":"The review reports more than 1,600 state AI bills proposed since 2019, with 77% categorized as controlling legislation; at least seven states with sandbox or pilot structures; and 12 city use cases across permitting, employee productivity, public safety, resident services, public engagement, and infrastructure analytics. Interviews characterized actual usage as pragmatic and modest, concentrated in drafting, translation, summarization, and basic automation, while cities reported shortages of technical talent, infrastructure, project-scoping capacity, and clear ROI.","architectureImplications":"State shared platforms can provide secure model access, common data services, evaluation, and reusable components, while cities may need narrower service integrations and external capacity partners. Architecture decisions must align with budget, staffing, data readiness, and the ability to operate and monitor systems after pilots end.","governanceImplications":"Pair guardrails with an affirmative portfolio strategy, use sandboxes to generate reusable evidence and operating standards, fund implementation capacity, and establish partnerships with universities and civic institutions. Track which pilots graduate, stop, or remain experimental and why.","securityPrivacyImplications":"Walled gardens reduce uncontrolled use only if identity, data boundaries, approved models, logging, evaluation, and incident response are centrally operated. Local partnerships require clear data stewardship, access, confidentiality, intellectual-property, and exit responsibilities.","caveats":"The review describes its sample as representative but not exhaustive. Much of the evidence is qualitative, many referenced projects predate publication, legislation counts do not measure implementation, and the 12 city cases were partly selected for scale ambitions and news coverage."}},{"version":"enrichment/2026-09-05T02:42:45.193Z/new-america-ai-government-field-review","resource":{"id":"new-america-ai-government-field-review","title":"Field review finds SLED experimentation broadening faster than capacity, strategy, and outcome evidence","organization":"New America","sector":"State and local government","geography":"United States","publishedAt":"August 31, 2026","publicationDate":"2026-08-31","eventDate":null,"sourceName":"AI in Government: A Field-Level Review","sourceLabel":"New America field review","sourceUrl":"https://www.newamerica.org/insights/making-ai-work-for-the-public/ai-in-government-a-field-level-review/","evidenceClass":"independent-research","outcomeClass":"mixed","topics":["knowledge-work","developers-agents","infrastructure","governance-procurement","accessibility-workforce","operating-model"],"finding":"New America's review combined more than 40 practitioner and expert interviews, pilot work, literature review, legislation analysis, and a field scan of state and city activity. It found rapidly expanding legislative and pilot activity, with states favoring enterprise sandboxes or walled gardens and cities favoring stand-alone service pilots, but local capacity, coherent strategy, trust, infrastructure, budgeting, and evidence of return remained major constraints.","sledRelevance":"This is a useful operating-model snapshot of why visible experimentation does not automatically become durable public capability. It also highlights the distinct paths of states and cities and the potential role of universities, professional associations, and shared institutions in filling capacity gaps.","evidence":"The review reports more than 1,600 state AI bills proposed since 2019, with 77% categorized as controlling legislation; at least seven states with sandbox or pilot structures; and 12 city use cases across permitting, employee productivity, public safety, resident services, public engagement, and infrastructure analytics. Interviews characterized actual usage as pragmatic and modest, concentrated in drafting, translation, summarization, and basic automation, while cities reported shortages of technical talent, infrastructure, project-scoping capacity, and clear ROI.","architectureImplications":"State shared platforms can provide secure model access, common data services, evaluation, and reusable components, while cities may need narrower service integrations and external capacity partners. Architecture decisions must align with budget, staffing, data readiness, and the ability to operate and monitor systems after pilots end.","governanceImplications":"Pair guardrails with an affirmative portfolio strategy, use sandboxes to generate reusable evidence and operating standards, fund implementation capacity, and establish partnerships with universities and civic institutions. Track which pilots graduate, stop, or remain experimental and why.","securityPrivacyImplications":"Walled gardens reduce uncontrolled use only if identity, data boundaries, approved models, logging, evaluation, and incident response are centrally operated. Local partnerships require clear data stewardship, access, confidentiality, intellectual-property, and exit responsibilities.","caveats":"The review describes its sample as representative but not exhaustive. Much of the evidence is qualitative, many referenced projects predate publication, legislation counts do not measure implementation, and the 12 city cases were partly selected for scale ambitions and news coverage.","streamIds":["state-government","local-government"],"roles":{"sales":"Interpretation — Problem and stakeholders: State and city CIOs, program owners, budget and procurement leaders, universities, and civic partners may have many pilots but insufficient sustaining capacity. Discovery: Which experiments have an operational owner, baseline, funding, and evidence for graduation or closure? Value hypothesis: Matching scope to local capacity could improve portfolio decisions and reduce unsupported expansion. Potential engagement: Review a bounded portfolio and assess one shared-platform or service-integration candidate. Evidence boundary: The qualitative review and legislation counts describe activity and experience. They do not establish causal return, the success rate of all pilots, or a universal preference for state sandboxes or city stand-alone tools. Transfer depends on staffing, data, and service conditions.","engineering":"Interpretation — Fit: Consider shared model/data services where states can operate common controls and narrower integrations where cities have limited capacity. Architecture: Reuse approved identity, data boundaries, model access, evaluation, logging, and incidents while keeping service accountability explicit. Prerequisites: Scoped workflow, data readiness, support skills, budget, and partner stewardship agreements. Constraints: A walled garden is controlled only if maintained; procurement and integration effort may dominate model choice. Security: Verify approved data paths, partner access, confidentiality, and exit arrangements. Proposed validation: Run representative tasks, measure quality and operational effort, and have the intended support team handle a failure and model change. Demonstrate maintainability rather than accepting a successful initial demonstration as production readiness.","delivery":"Interpretation — Work and dependencies: Record hypotheses, baselines, funding, ownership, dependencies, and graduate/stop/continue decisions in the portfolio. Ownership: Programs own mission outcomes; central technology and procurement provide shared controls; external partners have explicit data, access, intellectual-property, and exit duties. Skills and adoption: Include scoping, service design, support, and workforce training before access expands. Governance checkpoints: Review evidence and operating funds at graduation and capture lessons from stopped pilots. Proposed acceptance: The workflow has measured service outcomes, staffed support, tested incidents, and an evidence-based decision. Risks: Pilot counts and legislation can be mistaken for capability, partners can create dependency, and modest workflow benefits may not justify a platform local staff cannot sustain."},"retrievedAt":null,"enrichedAt":"2026-09-05T02:42:45.193Z","enrichmentBasis":"archived evidence"}}]}