AI-Powered Mobile Proctoring Frameworks using Machine Learning Algorithms in Higher Education: Post-Covid Trends, Challenges, and Ethical Implications

Abstract

The rapid transition to online learning during and after the COVID-19 (Corona Virus Disease) pandemic has heightened the need for secure, scalable, and ethical online exam systems. AI-powered mobile proctoring frameworks have emerged as viable alternatives to traditional invigilation methods, enabling automated anomaly detection and behavior analysis through machine learning algorithms. This systematic review examines post COVID trends, technological developments, challenges, and ethical implications of mobile AI proctoring in higher education. Following PRISMA 2020 guidelines, 180 studies were retrieved and screened, with 20 peer-reviewed articles meeting the inclusion criteria. Findings reveal that while AI-powered proctoring enhances scalability, integrity, and real-time monitoring, it raises significant concerns about privacy, algorithmic bias, accessibility, and technical reliability. The review identifies gaps in relation to technical and methodological issues, ethical and social concerns, and institutional and infrastructural readiness. This review illustrates a lapse in the existing literature, which focus on resource intensive proctoring frameworks without considering mobile compatibility and light-weight frameworks, discusses technical challenges, and recommends future research directions to balance technological effectiveness with ethical standards.