Faculty have used artificial intelligence (AI) to develop course content, assist with grading, and streamline administrative tasks. We extend these applications by demonstrating how a course-specific AI chatbot can audit educational materials. A large language model–based generative AI system (ChatGPT, v5.2) was configured for Animal Physiology I, the first course in a two-semester sequence for animal science graduate students and first-year veterinary students. The knowledge base was limited to course notes rather than unrestricted internet access, requiring the chatbot to operate within the same content boundaries as students. Furthermore, the system was instructed to function as a course-aligned assistant for graduate- and professional-level learners. The chatbot was then used to evaluate alignment between learning objectives and instructional materials, map quiz and exam questions to student learning objectives, and generate explanations for assessment items. A chatbot-assisted audit of the course module on male reproduction confirmed strong alignment between objectives and learning materials, suggesting a lack of curricular drift. Learning objectives spanned both lower- and higher-order cognitive domains; however, the chatbot suggested that some objectives classified as lower-order by the instructor may, in practice, require higher-order reasoning. All multiple-choice and matching questions from quizzes and exams mapped to at least one learning objective, with most aligning to multiple objectives. The chatbot generated accurate rationales for assessment items, explaining why each response option was correct or incorrect. More complex assignment questions—those requiring figure interpretation or integration of information across multiple resources—necessitated advanced prompting, highlighting areas where students may benefit from additional instructional scaffolding. These findings illustrate the potential of course-constrained AI chatbots to support instructional alignment, learning objective refinement, efficient formative feedback delivery, and identification of areas where students may require additional support. This approach provides a practical and scalable strategy for enhancing teaching effectiveness in upper-level, concept-dense science courses.