• About half of the states have released or endorsed guidance on AI in education; most were released in 2024.
• State leaders should continue to develop, monitor, and update AI guidance for local education leaders, while the federal government should continue to share research and best practices.

​Artificial Intelligence (AI) in education is already here and has been for years. Teachers use AI teaching assistants, and students consult chatbots to answer questions. How did we get here?

When ChatGPT was released in late 2022, a surge of questions arose regarding the role of generative AI—technology that can create or generate new content—in education. Should teachers prohibit the use of AI? Is that even feasible? How can we detect cheating or plagiarism? Moreover, how can these tools assist teachers with administrative tasks or in tailoring lesson plans for individual students? How can students use this technology to enhance their creativity or access personalized tutoring?

According to survey findings released last year, 79% of educators feel their districts are taking too long to create a policy on AI in education.

Luckily, the effort to create guidelines is spreading across the states, with various state education agencies or education groups publishing state-specific guidance to help local education leaders and teachers.

I recently wrote about how the state of Utah has been navigating this topic via public policy and its efforts to be a leader in this space.

How are the 50 states doing when it comes to creating guidance for AI in education? How are national and federal leaders helping? Here are some takeaways.

About half of the states have guidance on AI in education

According to one database and an internal review, 27 states offer some sort of guidance on AI. Some of these directives are official guidance from the state’s education agency, while others are recommendations from education groups reviewing state-specific issues that state officials endorse. Most are broad, covering general principles like potential opportunities and risks, best practices, ethical uses (like plagiarism issues), and data privacy. Having a broad approach makes sense because it’s a new area that is constantly evolving.

Though the topic of generative AI initially received a lot of public attention at the end of 2022, most states that have guidance have produced it recently; in fact, most were released during 2024. Leading the way were Oregon and California, which were the first two states in the nation to publish guidance (Oregon in August 2023 and California in September 2023).

The vast majority of guidance came out or was approved in mid- to late-2024. You can see a general timeline here with guidance being published by North Carolina in January 2024; Virginia in January 2024; Washington in January 2024; Connecticut in February 2024 (a framework for all state government published on this date though the education department website hosts many different resources); Ohio in February 2024; Indiana in April 2024; Mississippi in April 2024; Oklahoma in April 2024; Utah in April 2024; Arizona in May 2024; Hawaii in May 2024; West Virginia in May 2024; Alabama in June 2024; Delaware in June 2024; Kentucky in June 2024; New Jersey in June 2024; Wisconsin in June 2024; Minnesota in July 2024; North Dakota in July 2024; Wyoming in July 2024;  Colorado in August 2024; Louisiana in August 2024; and Michigan in December 2024.

So far in 2025, two states have published guidance: Georgia in January and Maine in February.

Of course, this means almost half of the states still lack state guidance, leaving many local leaders and educators to work out issues on their own or to seek resources to understand these issues from other leaders in this space, like TeachAI, The AI Education Project, Khanmigo (Khan Academy), or SchoolAI. Other states, like Illinois, Maryland, and Tennessee, are pushing legislation to jump start momentum in their states. Since this issue is not going away, we anticipate that more states will adopt or publish guidance in the coming months.

Federal guidance exists, but it could be a continuing effort

In May 2023, under the Biden administration, the U.S. Department of Education’s Office of Educational Technology published a document called “Artificial Intelligence and the Future of Teaching and Learning: Insights and Recommendations.” It includes helpful definitions, acronyms, and recommendations that may be a starting point for those looking for ideas.

Though not specific to education, the current Trump administration published a presidential action statement in January 2025 called “Removing Barriers to American Leadership in Artificial Intelligence,” which seeks to “solidify our position as the global leader in AI.” This goal, no doubt, will require education to use, adapt, and teach about AI.

Though the future of the U.S. Department of Education is unclear under the Trump administration, a legitimate function of the department is to research and share best practices, which it should continue doing as the issue evolves.

Conclusion

In short, state policymakers have work to do. Many states need to begin the process of thinking through guidance, and those that have it will need to be vigilant to keep up with the private sector, which will be advancing this issue faster than policymakers can. We hope policymakers’ approach is one that welcomes innovation, is slow to overregulate, and is prudent in focusing on key principles as new questions arise.

Sutherland Institute intern Alexis Morgan contributed research to this article.