Artificial intelligence (AI) is transforming virtually every industry, and the public sector cannot afford to fall behind. State lawmakers across the country have passed laws that create new opportunities to integrate AI tools into government services. Although roughly 90 percent of state technology offices have launched AI pilot programs, the gap between states that operate innovative, whole‑of‑government sandboxes and those that struggle to track outcomes or scale beyond isolated agency tests continues to widen. Without clear evaluation metrics, states risk repeating cycles of experimentation without meaningful implementation. Policymakers can close this gap by adopting emerging state models that help governments follow through on pilot programs to deliver measurable, evidence-based improvements to public services using AI.
Utah emerged as an early leader by establishing a statewide AI pilot program through the 2024 Artificial Intelligence Policy Act. The law created the Office of Artificial Intelligence and authorized it to run a regulatory sandbox that lets companies test AI tools under temporary regulatory relief and state supervision. Instead of restricting AI use upfront, Utah evaluates tools in real‑world conditions and builds regulation around evidence rather than assumptions. The health technology company Doctronic, for example, uses an approved AI system to help patients with chronic conditions renew prescriptions at participating pharmacies. Utah’s approach grounds future regulation in real performance data and gives the state a scalable way to evaluate AI systems.
In 2024, Connecticut established an AI Responsible Use framework that set statewide standards for safe and transparent AI deployment. As part of that framework, the state in 2025 created the AI Engagement and Enablement Lab, giving agencies a controlled environment to safely test AI deployments, supported by an AI Advisory Board that oversees implementation, coordinates with labor organizations, and engages private sector and academic experts. This structure has enabled 20 pilot programs, including Microsoft Copilot for productivity, automated Q&A tools for citizen inquiries, and systems that identify election‑related misinformation. Together, these elements ensure AI adoption happens with agency alignment, workforce coordination, and continuous expert oversight rather than fragmented experimentation.
In 2024, Ohio created an AI Governance Framework to provide statewide planning, implementation, procurement, and governance requirements for the use of AI. The framework created a multi-agency AI Council to govern the statewide use of AI and directed the Council to establish an emerging technology sandbox. The sandbox, which went into effect in March 2026, enables agencies to experiment with new AI tools in secure, controlled environments. It also supports more than a dozen agencies, all operating under state-mandated privacy requirements, including controls that ensure sensitive citizen data cannot resurface in public AI systems later on, safeguards that many vendors already offer but that Ohio now requires for all pilots. The Department of Administrative Services oversees the initiative and works with academic and legal experts to maintain ethical standards and workforce readiness. Ohio has already deployed pilots such as AI‑equipped vehicles that detect road defects and AI‑powered analysis of Medicare prior-authorization requests. This balance of innovation and strong safeguards gives agencies a safe, structured environment to test AI systems, strengthening their long-term operational capacity and public trust.
Texas enacted the Texas Responsible Artificial Intelligence Governance Act in 2025, which took effect in January 2026. The law establishes a comprehensive AI regulatory sandbox administered by the Department of Information Resources in coordination with a seven‑member AI Council, consisting of AI experts. The sandbox gives developers up to 36 months to test AI systems, from healthcare diagnostics to infrastructure management, under temporary regulatory relief. By offering a structured pilot environment, Texas encourages private sector innovation while collecting the data needed to shape long‑term, evidence‑based AI policy. This extended testing window also positions the state to evaluate more mature AI systems and build durable regulatory frameworks.
North Carolina Governor Josh Stein issued Executive Order (EO) No. 24 in 2025 to create an AI Accelerator within the Information Technology Department, setting up a 60‑day rapid‑testing cycle where agencies can pilot AI solutions in a secure environment. The EO also requires every state agency cabinet to form an AI Oversight Team to identify use cases. This framework has enabled North Carolina’s Treasury agency to pilot an AI system that streamlines management of public financial records. By pairing an internal testing hub with a high‑level AI Leadership Council, the state has built a repeatable pipeline for moving AI from proof‑of‑concept to statewide deployment.
These five states have moved beyond cautious observation by creating structured environments that prioritize hands‑on experimentation. Other states can follow their lead by moving away from siloed, agency-by-agency pilots that often operate in isolation and lack shared evaluation standards and toward centralized hubs for rapid internal deployment and multidisciplinary oversight councils to coordinate cross‑agency strategy. By shifting away from restrictive blanket bans and toward supervised, whole‑of‑government testing, states can generate the data needed to craft effective and safe guardrails. To ensure that public services remain modern and secure, policymakers should move from theory to practice and adopt these emerging sandbox models to close the growing gap between AI’s potential and its responsible integration.
Image credit: Generated with DALL-E
