Securing the metaverse: Machine learning–based perspectives on risk, trust, and governance

The rapid expansion of the metaverse presents significant cybersecurity and privacy challenges, requiring structured, data-driven analysis. This study applies the ADO-TCM framework and BERTopic modeling to examine drivers of cybersecurity risk, theoretical responses, and interdisciplinary research gaps. Using PRISMA guidelines, 86 peer-reviewed studies were analyzed to identify key antecedents—technological vulnerabilities, user behavior, regulatory fragmentation, economic incentives, and cultural factors—shaping decisions in compliance, deployment, and education. These, in turn, influence outcomes like trust, threat mitigation, and scalability. The review identifies five latent themes: secure identity, privacy, trust, governance, and AI’s role in shaping risk. The study maps diverse theoretical lenses—cognitive, behavioral, strategic, and technological—used to interpret immersive threats and decision-making in metaverse contexts. Contributing a novel, empirically grounded synthesis, this research advances the information management literature and proposes a forward-looking agenda focused on adaptive security, ethical AI, interoperability, regulatory convergence, and intelligent, user-centric architecture for immersive ecosystems.