The integration of artificial intelligence (AI) in education promises to revolutionize learning through personalized content, automated assessments, and administrative support. However, it also raises critical ethical concerns, particularly around fairness, bias, and the protection of students’ rights. The European Union’s AI Act, which came into effect in 2024, introduces a regulatory framework to address these challenges, especially in high-risk sectors like education. This paper explores the implications of the AI Act for educational institutions, focusing on its prohibition of emotion inference AI, its regulation of high-risk AI applications, and its requirements for transparency, data quality, and human oversight in high-stakes decision-making systems. It also examines various AI techniques for emotion inference, such as facial expression analysis and biometric measurements, highlighting ethical concerns related to privacy, accuracy, and bias. By analyzing the Act’s provisions, this paper discusses how these regulations aim to increase students’ privacy and mitigate biases related to race, gender, socioeconomic status, and other factors, while promoting ethical AI deployment in education. The paper concludes by discussing the responsibilities of educators and institutions in ensuring compliance with the Act, ultimately fostering a more equitable and accountable use of AI in educational settings.