Simple Explanation (ELI5)

Think of artificial intelligence as a personalized smartwatch that, instead of tracking your workouts, helps you study: it adapts to your pace, reminds you of forgotten topics, and challenges you with custom exercises. Over the past few years this ‘smartwatch’ has started running everywhere in schools and universities, but experts are already asking themselves ‘how do we use it without getting hurt?’.

Key Advantages & Current Limits

🟢 Pros: Intelligent tutoring systems and personalized learning paths tailor study experiences, especially in higher and medical education. 🟢 Pros: Growing ethical focus (68% of top-cited papers) pushes for clear rules on bias and privacy, lowering risks for students. 🔴 Cons: K-12 research accounts for less than 5% of the total, leaving a critical gap for primary and secondary schools. 🔴 Cons: The surge in interest (from dozens to over 1,200 publications) risks spawning untested applications or inconsistent ethical standards.

Future Outlook & Takeaways

In the next 3-5 years we’ll see AI tutors embedded directly in student apps, offering real-time answers and personalized feedback, while governments and companies roll out European and global rules ensuring transparency and privacy. At the same time there will be a push to bring these technologies into K-12 schools, filling the current gap, but strong collaboration between technologists, educators, and ethicists will be needed to avoid algorithmic discrimination.

Is Artificial Intelligence Revolutionizing Schools? Exploring...

Article

100%

In the last four years artificial intelligence in education (AIEd) has exploded like fireworks. By analysing almost 3,000 papers from 1990 to 2024, researchers found that publications jumped from a few dozen to over 600 in 2023, surpassing 1,200 by November 2024. The United States and China lead the race, with University of London and the University of California system at the core of the collaboration network.

Keywords tell the story: until 2020 “AI” and “machine learning” dominated, but from 2022 “ChatGPT” and “ethics” appear. This shift shows that, besides testing new technologies, the scientific community is increasingly worried about moral implications. In fact, 68 % of the most‑cited papers discuss ethical dilemmas, from data bias to student privacy.

The most common applications are in higher and medical education, where intelligent tutoring systems and personalised learning paths promise to make studying more tailored. Yet research on K‑12 remains scarce, leaving a crucial gap. The authors therefore call for stronger interdisciplinary collaboration and solid ethical guidelines, so AI can truly improve education without creating new problems.

Bonus

100%

The study employs an advanced bibliometric methodology, combining co‑citation, co‑authorship, and co‑keyword analyses using VOSviewer v1.6.20 and Python 3.11.5. Data were extracted from the Web of Science, filtering 2,800 English‑language publications and removing duplicates and irrelevant papers. Network analysis revealed three macro‑thematic clusters: (i) AI technologies (machine learning, deep learning, generative models), (ii) educational applications (personalised learning, intelligent tutoring systems, MOOCs), and (iii) ethical‑social issues (bias, privacy, governance). Temporal keyword mapping, performed with a TF‑IDF‑based clustering algorithm, shows a clear shift from a technical focus (2018‑2020) to a socio‑technical one (post‑2022). Betweenness centrality identified the most influential institutions, confirming the hub role of London and California. Citation analysis disclosed that 68 % of the most‑cited papers address ethical topics, indicating growing community awareness. Nevertheless, K‑12 studies account for less than 5 % of the total, highlighting a significant gap. The authors suggest expanding longitudinal datasets and integrating social impact metrics to better assess the long‑term consequences of AIEd.