AI Governance in Computing Education: Toward Discipline-Aware Academic Integrity Policies

Generative artificial intelligence (GenAI) systems, particularly large language models (LLMs), are rapidly transforming the technological environment in which higher education operates. Universities have responded with a wide range of policies, from strict prohibition to unrestricted adoption, often without considering how learning processes differ across academic disciplines. In this opinion piece, I argue that computing education represents a special case within the broader debate on AI in education. In my view, programming learning occurs within artifact-centered environments where students routinely interact with code repositories, libraries, and documentation. As a result, students develop expertise through the study, adaptation, and integration of existing computational artifacts. I therefore contend that GenAI expands an existing ecosystem of learning artifacts rather than introducing an entirely new form of assistance. Building on this observation, I propose a discipline-aware policy framework for AI use in computing education. This framework advances five principles for AI governance in programming courses: learning-centered policy design, transparency and disclosure of AI use, tiered assignment-level AI permissions, evidence-based integrity evaluation, and equitable access to AI tools. I further argue that effective policy cannot rely solely on institutional rules or AI-detection mechanisms. Instead, scalable educational infrastructure – including code provenance tracking, conversational code explanations, and AI-assisted oral checks – is necessary to make policy implementation operational in large and online computing courses. By bringing together insights from learning theory, computing education research, and institutional governance, I offer a forward-looking view of AI policy that aligns technological capabilities with pedagogical goals. Rather than treating GenAI solely as a threat to academic integrity, I frame AI governance as a design problem: ensuring that AI-assisted tools support student reasoning and conceptual engagement within modern programming learning environments.