Artificial intelligence (AI) is moving rapidly into K-12 classrooms. From tutoring tools to academic content generation, schools are experiencing AI across nearly every aspect of teaching and learning. This rapid expansion has created a new challenge for state leaders: how can we embrace the promise of educational innovation via AI while also establishing clear guardrails to protect students and preserve trust in public education?
In 2026, that challenge has become a major policy priority. The PIE Network tracked nearly 100 state bills this year that could directly affect students’ use of AI in K-12 education. That activity is unfolding alongside more than 1,500 AI-related bills introduced by lawmakers nationwide this year. Together, these numbers demonstrate that states are moving beyond early exploration of AI’s capabilities and into active governance.
As state policymakers explore regulation of AI, the more important question is whether emerging policies are simply reacting to immediate technological disruption or are truly building substantive frameworks for responsible adoption and implementation long-term.
ExcelinEd’s Guardrails for AI-Powered Educational Tools in K-12 Schools model policy offers one useful lens for assessing progress. It emphasizes clear expectations around the most important elements of privacy, transparency, safety and product design so states can encourage innovation without sacrificing student protection.
Across the country, several policy patterns are emerging. While approaches vary, many states are translating public concerns into action.
Safeguarding student information is a top priority. In California, proposed A.B. 1159 would prohibit student data from being used to train artificial intelligence models unless doing so directly benefits the school. The proposal reflects growing concern that student information should not be used to develop commercial products unrelated to educational purposes.
Alabama has taken a similar approach. State leaders released a procurement model policy that recommends a “data usage restriction” clause for district contracts, providing districts with greater leverage during vendor negotiations.
As AI-powered tools become more interactive, states are beginning to focus not only on what tools do, but also on how they are designed. In Oregon, enacted S.B. 1546 requires certain design features intended to protect minors, including measures that reduce excessive or compulsive use when a user is known or reasonably presumed to be a child. In Washington, enacted H.B. 2225 adds accountability through reporting requirements related to harmful behaviors. This helps ensure that risks are not only identified but documented and addressed.
AI adoption and use in schools will ultimately depend on educators. Virginia recognized this through enacted S.B. 394, which directs the state education agency to provide guidance that includes such areas as teacher training. Likewise, Maryland enacted S.B. 720, requiring the state to offer teachers professional development opportunities related to AI.
These steps matter because many districts do not have the capacity to build comprehensive implementation frameworks on their own. State-level support helps educators locally adopt tools responsibly.
States are also beginning to recognize that preparing students for the future requires much more than the regulation of new technologies. Several states are exploring how AI literacy can be integrated into academic standards and graduation requirements. Enacted bills in Utah for middle school, as well as proposed bills in Mississippi for high schoolers, reflect the understanding that students need both technical familiarity and personal judgment to navigate an increasingly AI-connected world.
Beyond these core themes, states are continuing to work through oversight questions. Many are leaning on local district implementation, as seen in Ohio. Others are placing greater emphasis on parental choice. In Oklahoma, proposals would require schools to notify parents when AI tools are used and provide opt-out opportunities in certain circumstances.
Additionally, while not integrated into policy, some states like California are exploring the use of outcomes-based contracts, holding vendors accountable for results by tying significant portions of their financial compensation to measurable student outcomes.
Even as states make meaningful progress, important gaps remain. Much of the current legislative activity has focused on basic protections, while signaling that a focus on AI policy matters. However, in a rapidly evolving landscape, states are balancing efforts to address the most visible risks associated with AI with the need to build rigorous frameworks that support long-term adoption, innovation and opportunity.
One of the most immediate challenges is that responsibility for evaluating AI tools is falling heavily on local districts. States may want to consider providing standardized evaluation frameworks, vetted procurement criteria or centralized libraries of approved tools. Such supports could help under-resourced districts navigate complex technical and legal questions and avoid fragmented decision-making and uneven student protections across localities.
Most current proposals do not require vendors to provide auditable records of user interactions, disclose how systems generate outputs or demonstrate that tools have been evaluated for bias, accuracy and reliability before being used by students. With stronger transparency expectations, school leaders could approach procurement decisions with the information necessary to assess risk.
The next phase of AI policy could move beyond isolated requirements toward more coherent implementation, and state education agencies can play an important role here.
First, states can develop “plug-and-play” procurement language for districts, building on the approach seen in Alabama. Clear model contract language can help local leaders address privacy, transparency and data-use concerns without building standards from the ground up.
Second, states can strengthen administrative guidance even in the absence of new legislation. Existing AI guidance can be updated to more explicitly address evaluation criteria, vendor disclosures, instructional design expectations and implementation planning. Maryland offers a useful example. Its guidance moves beyond high-level principles by identifying practical vendor considerations and local planning supports.
Finally, states can acknowledge that responsible AI adoption is not solely about managing risk but also about building capacity. The strongest policy frameworks will protect students, equip educators and help schools make the important decisions about when and how AI can improve teaching and learning.
In 2026, states have clearly entered the AI policy era. The opportunity now is to build from these early policy efforts toward frameworks that can support long-term governance and innovation.
Status of bills, enacted laws, and state guidance related to AI use in K–12 educational settings — as of May 2026.
