The integration of Artificial Intelligence (AI) in higher education is rapidly transforming traditional teaching methodologies. AI-powered tools, such as ChatGPT, have demonstrated significant potential in assisting students with problem-solving, conceptual understanding, and self-assessment. However, evaluating the reliability, accuracy, and pedagogical impact of these models in specific academic disciplines remains a crucial area of study.

In this work, we present an analysis based on an innovative teaching experience conducted in the Statistics course of the Aerospace Engineering Bachelor's Degree at the Polytechnic University of Valencia. The study focuses on assessing the capabilities of ChatGPT in solving statistical problems used in student self-assessments. These assessments cover various statistical topics and problem types commonly encountered in engineering education. By systematically providing these problem sets to ChatGPT and comparing its responses to expected solutions, we evaluate the model’s effectiveness, limitations, and potential as an educational tool.

The analysis explores multiple dimensions, including:
(1) ChatGPT’s accuracy in solving statistical exercises across different topics,
(2) its performance in handling different question formats, such as numerical problem-solving, multiple-choice reasoning, etc.
(3) a comparative assessment between ChatGPT’s results and student performance, and
(4) the role of AI as an instructional assistant for students preparing for statistical examinations.

Additionally, we examine whether the AI-generated solutions align with the problem-solving approaches expected in engineering contexts and discuss how such tools might influence students’ learning strategies.

The study also reflects on the pedagogical implications of AI-driven assistance in statistics education, raising questions about its potential for personalized learning, the risks of over-reliance, and the evolving role of educators in an AI-enhanced academic landscape. By integrating empirical data from student assessments, ChatGPT-generated responses, and qualitative insights from student feedback, we aim to contribute to the ongoing discourse on AI’s impact in university-level STEM education.

Our findings will provide valuable insights into the advantages and challenges of using AI in statistics teaching, supporting the development of more effective strategies for incorporating AI-driven tools into engineering education.

Keywords: Artificial Intelligence, ChatGPT, Statistics Education, Critical thinking, Self-Assessment, Aerospace Engineering, Higher Education Innovation.