Mapping the landscape of Artificial Intelligence in education

In a context of increasing digitalisation, Artificial Intelligence (AI) is establishing itself as a technology with apparent potential to transform teaching and learning processes. This research analyses the main trends, applications and emerging challenges of AI in the field of education in Ibero-American countries through a bibliometric and content analysis. Using a mixed methodology, developed with the PRISMA protocol, and focusing on empirical studies published between 2020 and 2024 in the Scopus database, 215 articles addressing the use of AI at various levels and in various areas of education are analysed. The results show a significant increase in scientific production, with Spain as the most represented country, and a strong tendency to publish in English. Collaboration between authors and the concentration of publications in a small number of journals predominate. In terms of content, tools such as ChatGPT, Fliki AI and Leonardo AI stand out, used to generate educational content, facilitate adaptive learning and automate assessment. However, certain risks are also identified, such as technological dependence, lack of critical thinking, algorithmic biases and the growing digital divide, especially in contexts with limited infrastructure or poor teacher training. This study highlights the need to integrate AI into education with caution, based on ethics and pedagogy. Solid teacher training and a critical and prudent approach seem to remain the pillars for ensuring that AI complements rather than replaces the role of teachers, promoting inclusive and quality education.