Artificial intelligence applied to the early detection of psychoeducational needs in early childhood and basic Education students

Abstract

The integration of artificial intelligence into educational systems has created new possibilities for identifying learning patterns, academic difficulties, and individual student characteristics. This article analyzes the opportunities, challenges, and perspectives associated with using artificial intelligence tools for the early detection of psycho-pedagogical needs in early childhood and basic education students, as well as their potential to foster personalized learning. A review of recent scientific literature was conducted, focusing on systematic reviews and studies published between 2024 and 2026 in peer-reviewed academic journals. The evidence indicates that intelligent learning systems, adaptive learning, learning analytics, machine learning, and certain pattern recognition systems can help identify indicators linked to learning difficulties, language development, academic performance, and other educational needs. However, the evidence also reveals limitations regarding data quality and representativeness, algorithm explainability, child privacy, bias, teacher preparedness, and a scarcity of longitudinal studies conducted in real-world educational settings. The study concludes that artificial intelligence can serve as a complementary tool to strengthen observation, assessment, and psycho-pedagogical decision-making processes, but it should not replace professional judgment or comprehensive assessment procedures. Its greatest potential lies in supporting educational systems capable of detecting early warning signs, guiding differentiated interventions, and providing learning experiences better tailored to each student's characteristics.