AI in Schools: 80% of Students Use It, 6% Have Clear Policies

01A Widening Gap Between Adoption and Policy

The 2026 Artificial Intelligence Index Report, published on April 13 by Stanford University's Institute for Human-Centered AI (Stanford HAI), lays bare a critical gap in K-12 education: four out of five high school and college students now use artificial intelligence tools for school-related tasks, yet only 6% of teachers report that their schools have clear policies on the subject. The current challenge is not rooted in technology adoption — it is rooted in profound institutional unpreparedness. The report, spanning 423 pages across nine chapters, is one of the most comprehensive independent assessments of the global AI landscape. Its education chapter reveals that generative AI reached 53% of the world's population within just three years — faster than the personal computer or the internet. Outside the classroom, AI skills are growing rapidly. Inside it, formal education has yet to find a coherent path forward.

02The Cost of Institutional Silence

The absence of clear guidelines carries measurable consequences for student learning and well-being. Recent data indicates that 20% of students' in-school interactions with artificial intelligence involve troubling behaviors, including cheating, bullying, or self-harm. Research conducted by the University of Southern California found that more than a third of teenagers believe AI use is worsening their ability to think independently — and that most parents are unaware of how deeply their children rely on these tools. Morgan Polikoff, a professor at USC's Rossier School of Education, frames the core tension clearly: learning is rarely about the final product; it is almost always about the process of getting there. When artificial intelligence shortcuts the effort required to master new skills, the long-term consequences for individuals and society may be substantial. The available evidence points to an urgent need for policies that guide both teachers and students on appropriate and inappropriate uses of the technology.

03Brazil's Response and the MEC Framework

Brazil's student adoption of AI mirrors the global trend. Data from the TIC Educação survey shows that seven out of ten high school students admit to using AI tools for their studies. In response, Brazil's Ministry of Education (MEC) published the National Guiding Document on the Use of Artificial Intelligence for Teachers in Basic Education in April 2026, developed in collaboration with UNESCO and Brazilian university experts. The document proposes a four-pillar framework for teacher professional development: critical understanding of the technology, intentional pedagogical use, protection of student rights and well-being, and continuous professional training. A central distinction in the framework separates "teaching about AI" from "teaching with AI," requiring that any adoption of technological tools be accompanied by digital literacy instruction. Starting in 2026, curriculum updates aligned with these guidelines become mandatory for school networks and are tied to federal Fundeb funding transfers. The Stanford report itself notes that Brazil ranks second globally in producing information and communications technology graduates — behind only the United States. That technical pipeline is a meaningful asset, but it underscores the parallel urgency of preparing teachers, not just students, for the pedagogical dimensions of AI in the classroom.

04What to Watch Next

The evidence from Stanford's report and Brazil's new MEC framework reinforces a central premise: technology alone does not guarantee educational excellence. Meaningful progress occurs when advanced tools are mediated by prepared educators and supported by solid institutional policy. The focus must shift from outright prohibition or uncritical adoption toward building a culture of collaboration, data literacy, and intentional use. The immediate challenge for school administrators and policymakers is not to stop the use of artificial intelligence, but to integrate it in ways that advance equity and deep learning. The open question is whether schools can implement the proposed frameworks with the same speed at which students are already adopting new technologies — or whether the gap between use and governance will continue to widen.