A Transparent and Adaptive AI Assistant for Teaching Knowledge Engineering

As generative AI tools become more widespread, students are increasingly using them for assistance with complex tasks such as modeling ontology constraints. While the success of large language models have been widely explored, in-use applications remain underdeveloped, and experimental findings are often inaccessible to students or novice engineers. As a result, learners do not fully benefit from AI-assisted support or fail to critically engage with AI generated outputs. To bridge this gap, we propose a transparent, research-informed AI Assistant framework that follows hybrid intelligence principles and aims to support Knowledge Engineering education, with a focus on modeling logical ontology constraints. Preliminary results suggest that such a system can improve the accuracy of student-generated ontology models by over 10%.