AI Challenges in Higher Education: A Scoping Review of Ethical, Transparency, and Social Responsibility Issues

Although Artificial Intelligence (AI) holds potential in the field of education, it is important to address the ethical and pedagogical concerns surrounding its use. This study therefore aims to analyse the AI challenges in higher education, focusing in greater detail on ethics and regulation; transparency and data management; and social responsibility. Furthermore, a systematic review was conducted following the PRISMA methodology, covering articles from various databases spanning the period from before 2017 to 2024. Studies were selected based on specific inclusion criteria; of these, 49 met the study’s criteria and were categorised into eight main AI challenges: a) risk of exclusion due to technical barriers, b) reinforcement of discrimination through biased algorithms, c) threats to data privacy, d) potential displacement of human labour by automation, e) injustices in academic monitoring due to automated supervision, f) plagiarism and academic dishonesty, g) negative emotions and loss of human connection, and h) inappropriate and inaccurate content. In this regard, addressing these challenges leads to conclusions regarding practices such as the periodic review of algorithms, the implementation of evaluation strategies to regulate all forms of unethical use of AI, and a collaborative educational training programme on ethics and digital wellbeing for all academic users.