The Ministry of Education (MEC) officially released on Tuesday (10th) the national standards regulating the use of artificial intelligence (AI) in the Brazilian educational system. These rules apply from early childhood education to higher education, defining clear limits for the use of technology by students, teachers, and educational institutions.
One of the significant changes compared to the previous opinion, approved by the National Council of Education (CNE) in September, is the permission for AI tools to assist in grading essay questions. However, the assignment of the final grade remains the responsibility of the teachers.
The document's rapporteur explained that the previous version did not specify whether the use of AI in this type of assessment was prohibited. The modification aims to allow technology to function as support, provided there is constant human oversight.
The guidelines also stipulate that automated systems must never fully replace the function of education professionals or assume responsibility for pedagogical decisions. The rules vary according to the age group of the students and the level of education.
In early childhood education, children cannot use AI systems, even with teacher assistance. Educators, in turn, can use these tools for planning, pedagogical management, and accessibility purposes, provided the goal is to remove learning obstacles.
For elementary school in the initial years (1st to 5th grade), the use of AI is restricted to pedagogical activities that are planned, conducted, and supervised by education professionals. Students are prohibited from accessing these systems autonomously or without supervision, including interaction with conversational or generative tools, even when teachers are present. An exception exists for assistive technologies designed to overcome learning barriers, following specific criteria.
In the final years of elementary school (6th to 9th grade), generative AI tools can be used in pedagogical activities mediated by educators. However, individual, autonomous, or unsupervised student access remains prohibited. The document also provides for the gradual inclusion of AI education in the curriculum, alongside topics of digital, media, and computational literacy, aiming to equip students to understand how these tools work, recognize biases, fact-check information, and distinguish technological assistance from their own intellectual production.
At the higher education level, institutions are permitted to apply AI in teaching, research, and extension activities. Technology can also be a subject of academic study and investigation. Universities and other entities must create their own policies to regulate the use of these tools, integrating them into pedagogical projects, regulations, and institutional development plans.
A crucial aspect is transparency: institutions must specify when students and researchers need to declare the use of AI in their work. In scientific research, for example, its use must be reported if the tool's contribution is relevant to text writing, data analysis, literature review, programming, or other phases of the investigation. The guidelines are clear in stating that AI systems cannot be considered authors of academic or scientific works; the responsibility for the content and results always rests with students, teachers, and researchers.
In terms of university administration, automated tools can assist in planning, indicator analysis, and detecting dropout risks. However, decisions that significantly affect students' rights or academic trajectory cannot be made solely by AI systems. Finally, teacher training courses must incorporate content on digital literacy, pedagogical application of emerging technologies, critical analysis of educational data, and the ethical implications of AI.
This approval comes at a time when AI is already part of the routine of many Brazilian students. According to the 15th edition of the TIC Education survey, 37% of basic education students reported using generative AI tools in school research, but only 19% mentioned receiving any guidance from teachers on the use of the technology. With these new rules, the MEC seeks to standardize practices that already occur in schools and universities, balancing the benefits of AI with the need for human supervision, data protection, and pedagogical responsibility.
