University Students and Artificial Intelligence: Adoption, Learning, Integrity, Governance in Japan, the United States, Europe

Ninety-five per cent of UK undergraduates now use artificial intelligence. In Japan the figure reached ninety-two per cent in two years. Almost no university has caught up. University Students and Artificial Intelligence is a comparative, mixed-methods study of what students actually do with generative AI in Japan, the United States, and Europe - and what their institutions are doing about it. Drawing on twenty national and international student surveys, seven meta-analyses, randomised experimental evidence, freedom-of-information data on academic misconduct, and the EU Artificial Intelligence Act, it moves past the exhausted argument about cheating to the harder questions of learning value, authorship, equity, and institutional legitimacy. The book answers questions most commentary avoids. Does AI use actually improve learning? (The evidence is weaker than almost everyone citing it realises - and the field's most-quoted study was retracted.) Can detection software support a misconduct finding? (It produced a sixty-one per cent false positive rate against writers who learned English as a second language.) What does the EU AI Act really require of universities? (Far less, and far more specifically, than the commentary suggests.) Every statistic is reported with its population, sample size, and fieldwork period, and was checked against the organisation that published it. Widely circulated figures that failed that check are named. Two appendices give readers the entire evidence base in one table and an eight-question protocol for judging new statistics as they appear. The result is a practical, evidence-led framework - CLEAR - for university leaders, faculty, academic-integrity officers, education policymakers, and graduate students in education, law, and public policy. Zahir Hasan, MPPA, LLM, EdD, teaches at a university in Tokyo and writes on public policy, law, education, and institutional governance.