The Integration of AI Genes in Education: A Systematic Review of Opportunities and Challenges

Abstract

The rapid diffusion of generative artificial intelligence (GenAI) tools such as ChatGPT, Google Gemini, and Microsoft Copilot has generated unprecedented interest in their integration within educational settings, ranging from primary schooling to postgraduate professional training. This systematic review synthesizes evidence from twenty-five peer-reviewed studies published between 2021 and 2026 to map the opportunities and challenges associated with the integration of Generative AI in education. Following PRISMA 2020 guidelines, records were retrieved from Scopus, Web of Science, ERIC, IEEE Xplore, and Google Scholar, screened for relevance, and appraised for methodological quality. The review identifies five major opportunity domains: personalized and adaptive learning, automated assessment and feedback, content and curriculum generation, accessibility and inclusion, and support for research and professional development. Conversely, five recurring challenge domains emerge: academic integrity and misuse, algorithmic bias and equity gaps, over-reliance and erosion of critical thinking, data privacy and governance uncertainty, and uneven institutional and teacher readiness. Unlike earlier reviews that treat opportunities and challenges as parallel and static lists, this review introduces an integrative maturity-readiness framework that links each opportunity to the specific institutional safeguard required to realize it responsibly. The findings offer evidence-based implications for policymakers, teacher educators, and institutional leaders, while identifying longitudinal outcomes and equity-centered implementation as priorities for future research.