Abstract

The integration of Artificial Intelligence (AI) in education has transformed learning personalization and student performance management. This study aimed to analyze the technologies, implementation strategies, and reported outcomes in recent scientific literature regarding AI workflows and intelligent educational agents. Under the PRISMA methodology, a systematic review was conducted on 70 studies published between 2020 and 2026 in high-impact databases. The findings reveal a growing trend toward the use of learning analytics, multi-agent systems, and academic chatbots—tools that have proven effective in early risk detection and automated monitoring. However, critical barriers were identified, such as the lack of technological interoperability and ethical dilemmas in data governance. It is concluded that while these technologies are promising for the evolution of digital environments, their success depends on developing standardized architectures and conducting longitudinal studies to validate their long-term pedagogical impact.