Generative Intelligent Tutoring Systems (ITSs), powered by advanced language models like GPT-4, represent a transformative approach to personalized education through real-time adaptability, dynamic content generation, and interactive learning. This study presents a modular framework for designing and evaluating such systems, leveraging GPT-4’s capabilities to enable Socratic-style interactions and personalized feedback. A pilot implementation, the Socratic Playground for Learning (SPL), was tested with 30 undergraduate students, focusing on foundational English skills. The results showed significant improvements in vocabulary, grammar, and sentence construction, alongside high levels of engagement, adaptivity, and satisfaction. The framework employs lightweight JSON structures to ensure scalability and versatility across diverse educational contexts. Despite its promise, challenges such as computational demands and content validation highlight the main areas for future refinement. This research establishes a foundational approach for advancing Generative ITSs, offering key insights into personalized learning and the broader potential of Generative AI in education.

Although there are studies on digital competence in higher education, there are very few studies on digital safety competence. This study aims to explore the impact of digital safety competence on students’ higher-order thinking and AI-related outcomes. Using a cross-sectional design, 159 university students completed an online questionnaire to measure cognitive competence, Al self-efficacy, AI ethics, and moral competence. Results showed that digital safety competence was positively and significantly related to cognitive competence, AI self-efficacy, AI ethics, and moral competence (p < 0.05). Our study extends the literature by highlighting the role of digital safety competence. Educators and university policy makers may consider incorporating digital competence, especially in the area of safety, into their teaching and learning strategies.

Intelligent Tutoring Systems (ITSs) have significantly enhanced adult literacy training, a key factor for societal participation, employment opportunities, and lifelong learning. Our study investigates the application of advanced AI models, including Large Language Models (LLMs) like GPT-4, for predicting learning performance in adult literacy programs in ITSs. This research is motivated by the potential of LLMs to predict learning performance based on its inherent reasoning and computational capabilities. By using reading comprehension datasets from the ITS, AutoTutor, we evaluate the predictive capabilities of GPT-4 versus traditional machine learning methods in predicting learning performance through five-fold cross-validation techniques. Our findings show that the GPT-4 presents the competitive predictive abilities with traditional machine learning methods such as Bayesian Knowledge Tracing, Performance Factor Analysis, Sparse Factor Analysis Lite (SPARFA-Lite), tensor factorization and eXtreme Gradient Boosting (XGBoost). While XGBoost (trained on local machine) outperforms GPT-4 in predictive accuracy, GPT-4-selected XGBoost and its subsequent tuning on the GPT-4 platform demonstrates superior performance compared to local machine execution. Moreover, our investigation into hyper-parameter tuning by GPT-4 versus grid-search suggests comparable performance, albeit with less stability in the automated approach, using XGBoost as the case study. Our study contributes to the field by highlighting the potential of integrating LLMs with traditional machine learning models to enhance predictive accuracy and personalize adult literacy education, setting a foundation for future research in applying LLMs within ITSs.