As of mid-2025, one might assume that with artificial intelligence now writing essays, generating lesson plans, and reshaping how students interact with knowledge, the education world would have a unified strategy for figuring it out. But there’s no binding global framework, international consensus, and no central referee overseeing how AI is used in classrooms. Instead, we’re navigating a constellation of voluntary principles, hopeful white papers, and policy pilots.
The OECD’s AI Principles, first introduced in 2019, and UNESCO’s Recommendation on the Ethics of Artificial Intelligence offer strong normative guidance. They emphasize fairness, accountability, transparency, and ethical stewardship—especially in educational contexts. But they’re non-binding, relying more on international goodwill than regulatory force. The European Union’s forthcoming AI Act does promise enforceable protections, classifying education-related AI tools as “high-risk” and subject to clear oversight. Unfortunately, such legal rigor is still the exception, not the rule.
The reason? Global priorities diverge. The U.S. emphasizes innovation and market-driven AI growth; China embeds AI deeply into state-run systems and early education. Legal traditions, political structures, and cultural norms only widen the gulf. And since there’s no global enforcement body, even the most well-crafted policies often remain theoretical.
Still, something’s happening. In the United States, President Biden pushed for nationwide AI literacy from K–12 onward—an effort to make sure students understand not just how to use AI, but how it works. In parallel, President Trump’s administration made headlines in March 2025 with an executive order directing the Secretary of Education to “facilitate the closure of the Department of Education and return authority over education to the States and local communities.” The rationale? Reducing federal involvement in education.
The juxtaposition’s hard to miss. On one hand, the administration is promoting national-scale integration of AI in schools—through new grants, teacher training, public-private partnerships, and a Presidential AI Challenge for students and educators. On the other, it’s simultaneously seeking to dismantle the very federal department that would coordinate, implement, and evaluate those efforts. It's like designing a state-of-the-art ship while sawing off the rudder mid-voyage.
Meanwhile, New York’s showing what a more grounded and consistent approach could look like. The State Assembly introduced legislation to embed AI literacy into K–12 curricula and launched a $5 million initiative to establish Departments of AI and Society across SUNY campuses. New York City, having lifted its early ban on ChatGPT, is now equipping teachers with AI toolkits and piloting programs like YourwAI to help generate lesson plans. These initiatives are imperfect, but they’re proactive, pragmatic, and—most importantly—happening.
Canada’s still struggling with coherence. While several federal initiatives show promise—the Artificial Intelligence and Data Act (AIDA), the AI Strategy for the Federal Public Service, and the Canadian Artificial Intelligence Safety Institute (CAISI)—there’s no national policy on AI in education. With education a provincial responsibility, institutions are left to interpret and implement AI use independently. Some are embracing it; others are banning it outright. The result is a patchwork of uncoordinated responses.
Ontario’s Trustworthy AI Framework and the Higher Education Strategy Associates (HESA) AI Observatory offer some structure, and the Canadian Teachers’ Federation is actively calling for a unified K–12 approach rooted in equity and safety. Still, without a national coordinating body, much of this work exists in silos.
Globally and nationally, there’s momentum—but little cohesion. The conversations are getting louder, but the action still lags behind the technology. People are paying attention, but it’s mostly talk. If we want AI to enhance education, more than white papers, toolkits, and AI-themed competitions are necessary. The next generation of students, while pausing their video games, are asking generative AI to write their book reports, summarize them, and run them through Notepad to clear the data-trail, in elementary school. This while policymakers are still debating whether education policy should even be centralized. We may already have the answer.
