Towards an AI-Enabled Digital Wellbeing Assessment Framework: A Multidimensional Evaluation of Existing Digital Wellbeing Tools

Abstract

The rapid integration of smartphones, social media, online services, artificial intelligence, and other digital technologies into everyday life has increased the importance of understanding digital wellbeing. In response, numerous digital wellbeing applications and built-in device features have been developed to monitor technology use, reduce distraction, encourage healthier habits, and provide wellbeing support. However, existing tools differ considerably in their objectives, functionality, automation, personalization, and approaches to wellbeing assessment. More importantly, monitoring digital behaviour through screen time, application usage, notifications, or device activity does not necessarily provide a comprehensive assessment of an individual’s digital wellbeing. This study presents an exploratory comparative evaluation of existing digital wellbeing tools with particular attention to their monitoring capabilities, intervention mechanisms, personalization, artificial intelligence (AI), and relationship to wellbeing assessment. Ten representative tools were examined using publicly available information, official documentation, application-store information, and exploratory hands-on observation. The findings indicate that existing tools are predominantly specialized, with many concentrating on usage monitoring, application blocking, focus management, productivity, or individual aspects of mental wellbeing. AI capabilities are emerging but remain fragmented, while many interventions continue to rely on predefined rules, timers, thresholds, and user configuration. The review further identifies an important measurement gap. Although digital wellbeing has increasingly been recognized as a multidimensional construct, existing measurement approaches vary in theoretical foundations, dimensions, populations, and psychometric structures. Recent studies have begun developing and validating dedicated digital wellbeing instruments, but the literature continues to indicate a need for integrated, reliable, context-sensitive, and broadly validated approaches to digital wellbeing assessment. The study therefore argues that future AI-enabled digital wellbeing technologies require a stronger measurement foundation capable of distinguishing digital behaviour from broader wellbeing outcomes. The proposed direction is not a specific index or questionnaire, but a research pathway toward systematic measurement development that can support assessment, personalization, intervention evaluation, and responsible AI-enabled digital wellbeing systems.