📄 Abstract
Employers increasingly deploy artificial-intelligence systems that read employee sentiment from written communication, infer exhaustion from calendar and workload digital exhaust, and flag individuals as at elevated risk of burnout — tools marketed under the language of care rather than control. This paper develops a conceptual model of how AI-mediated wellbeing monitoring reshapes psychological safety, the shared belief that a work environment is safe for interpersonal risk-taking on which voice, dissent and error reporting depend. Integrating psychological-safety theory, the algorithmic-management and workplace-surveillance literatures, and organisational voice research, the paper argues that wellbeing-monitoring intensity depresses psychological safety principally through a mediating perception of surveillance threat, and that this relationship is not fixed but is conditioned by the transparency of the monitoring system, the degree to which employees attribute developmental rather than punitive intent to it, and whether a human manager mediates between algorithmic output and managerial action. Five propositions specify these relationships, including a boundary condition for India's IT-BPM sector, where the Digital Personal Data Protection Act, 2023 supplies a general consent-and-purpose-limitation framework but no sector-specific rule for wellbeing analytics, leaving adoption norms to be set informally inside a high-attrition, always-on operating environment. The paper closes with implications for the design of psychologically safe monitoring architectures and an agenda for empirical validation.
🏷️ Keywords
📚 How to Cite:
Dr. N Subbukrishna Sastry, Dr. Manjula Mallya M , AI-MEDIATED WELLBEING SURVEILLANCE AND THE BOUNDARY OF PSYCHOLOGICAL SAFETY , Volume 13 , Issue 10, October 2026, EPRA International Journal of Environmental Economics, Commerce and Educational Management(ECEM) , Pages: 22 - 35 ,