Integrating Generative AI for Personalized Learning in Cryptography Education

This chapter presents a longitudinal case study on the integration of generative artificial intelligence (AI), specifically Large Language Models (LLMs), into the teaching of cryptography at the undergraduate level. Conducted over eight academic terms at the Universidad de Guadalajara, the study analyzes how LLM-driven activities—such as personalized explanations, automated problem generation, and cryptanalysis simulations—impacted student learning. The implementation of a standardized final examination beginning in 2024B enabled consistent cross-cohort comparisons. Results show an initial performance dip during the transition phase, followed by stabilization and improved consistency. The findings suggest that LLMs can serve as adaptive scaffolding tools, supporting diverse learners in mastering abstract technical content. The chapter also explores the ethical, legal, and pedagogical considerations of using AI in education, offering a replicable model for integrating generative AI into STEM curricula.