Analysis of The Performance of Artificial Intelligence-Based Language Models in Electrical and Electronics Education Exam Questions

Over the last decade, advances in artificial intelligence have begun to transform educational practices. With the development of OpenAI's ChatGPT, it has become clear that AI chatbots' capabilities—such as generating human-like text, facilitating automated conversations, and answering questions—will have widespread impact across various sectors, including education and healthcare. Consequently, the use of chatbots in various applications has accelerated. Today's most popular chatbots, OpenAI ChatGPT and large language models (LLMs) like Google Gemini and Bard, are making significant progress with applications in various fields, including electrical and electronics education. In this study, the performance of ChatGPT (GPT-3.5), ChatGPT (GPT-5.1), Bard and Gemini was analyzed in answering 240 carefully prepared questions in Turkish language using clear, specific and plain text formats to determine knowledge levels, covering 12 sub-specialties in the field of electrical and electronics. Two experts analyzed the accuracy metric in the performance analysis, and the results are presented comparatively. The statistical significance of the findings was also determined, and it was observed that there were significant differences for the exam evaluations of some courses.