Based on Personalized Content: Addressing Homogenization in AI-Supported Learning Through the Smart Mistake Solver
Year: 2026
Type: article
Abstract: The adoption of artificial intelligence allows for an advancement in personalized learning, since the development of diverse types of learning materials and content may not succeed in always developing highly personalized experience. This article will explore personalized learning in relation to the homogeneity of a course program design based on an example of the Smart Mistake Solver program, an artificial intelligence learning assistance tool, which detects and corrects students' mistakes while dealing with math problems. The process of error detection, student model created with the help of Bayesian Knowledge Tracing (BKT) and development of personalized content are crucial aspects of the artificial intelligence support in learning. The conceptual analysis of Smart Mistake Solver demonstrates the limitations of artificial intelligence in enhancing personalized learning, which are conditioned by the necessity to use a list of predestined errors and measurable performance metrics. In conclusion, it was found that meaningful personalization should occur with respect to the specific needs of each learner within a learning experience along with distinction of learning materials; these cannot be fully met in all cases with the help of artificial intelligence.
Author: Zitong Fu
Institution: University of Washington
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7225740980