Four more states require their school districts to have an AI policy

Idaho, Maryland, Oklahoma and Virginia legislate. Oklahoma prohibits AI from justifying a disciplinary or evaluation decision on its own.

Four U.S. states — Idaho, Maryland, Oklahoma and Virginia — now require by law that their Education Department publish guidelines on safe and responsible use of artificial intelligence, and that each school board in turn adopt a policy aligned with this framework.

The texts don't look the same. Maryland requested the appointment of a coordinator to work with the state on a "productive and ethical" use. Oklahoma goes the farthest on safeguards: AI cannot serve as the main basis for an assessment, disciplinary sanction or any decision at high risk, parents can remove their child from a device without academic penalty, and districts must publish annually a list of tools used. Virginia, for its part, is opening a pilot funding program for AI applied to education, tutoring and teaching support.

These laws are in addition to an already dense landscape: by October 2025, thirty-four states had published guidelines on AI in schools. The gap with the reality of classrooms remains considerable — a May 2026 survey indicated that 82% of teachers had not received any formal instructions on the use of these tools.