Editorial summary. This is our text summary of an article published by sajhrm. Charts, figures, and the author’s full voice are at the original — read it there .
Editorial verdict
Methodologically transparent study with a coherent theoretical contribution, but cross-sectional self-report data from a single generational cohort in one country limits causal claims — the cascade model is plausible and internally consistent, but requires longitudinal and multi-context replication before organisational inferences are drawn.
Executive summary
This study examines the psychological mechanisms through which AI-enabled workplace monitoring translates into turnover intention (TI) among Generation Z employees in Indonesian organisations. The authors argue that AI monitoring differs qualitatively from earlier electronic performance monitoring (EPM) through four features — opacity, inference, automation, and permanence — that collectively expand perceived organisational surveillance beyond what employees consider legitimate. Using a cross-sectional survey of 390 participants analysed via partial least squares structural equation modelling (PLS-SEM), the study tests a 'boundary–autonomy cascade' model positing that AI monitoring raises perceived privacy intrusion (PPI), which activates psychological reactance, which in turn drives TI. All ten hypotheses were supported. AI monitoring related most strongly to PPI (β = 0.552), reactance was the strongest predictor of TI (β = 0.526), and the serial indirect path was significant (β = 0.126). The model explained 51% of variance in TI. The authors conclude that reactance — not privacy intrusion — is the proximal driver of exit cognition, and that technical privacy interventions alone are insufficient without addressing perceived autonomy and procedural legitimacy.
Key insights
- 1Psychological reactance, not perceived privacy intrusion, is the most proximal predictor of turnover intention (β = 0.526), indicating that the motivational response to autonomy threat is a stronger exit driver than the informational appraisal of boundary violation.
- 2AI monitoring differs from traditional EPM through four specific features — opacity, inference, automation, and permanence — which together produce sustained autonomy frustration that conventional EPM research does not capture.
- 3The serial cascade (AI monitoring → privacy intrusion → reactance → turnover intention) explains 51% of variance in TI, suggesting that digitally governed workplaces face a multi-stage psychological pathway to attrition that single-variable interventions are unlikely to disrupt.
Practical takeaways
- Organisations deploying AI monitoring systems face a cascading attrition risk that originates in perceived boundary overreach and culminates in reactance-driven exit cognition — addressing only the privacy layer while leaving opacity and autonomy restriction intact is unlikely to reduce TI.
- Procedural transparency, employee voice mechanisms, explainability features, and opportunities for discretion within AI-monitored workflows are identified by the data as the levers relevant to the proximal driver of exit, rather than technical privacy controls alone.
Frameworks mentioned
Psychological Reactance Theory
Brehm's (1966) theory positing that perceived threats to behavioural freedom produce a motivational state directed at restoring that freedom; applied here to explain how perceived privacy intrusion converts into the intention to leave.
Self-Determination Theory
Deci and Ryan's framework identifying autonomy as a basic psychological need whose frustration predicts disengagement and withdrawal; used to justify the centrality of autonomy in the cascade model.
Communication Privacy Management Theory
Petronio's theory holding that individuals manage privacy boundaries according to contextual norms, and that violations of those boundaries produce defensive arousal; applied to explain how AI monitoring triggers perceived privacy intrusion.
Contextual Integrity
Nissenbaum's (2004) framework defining privacy harm as information flows that exceed norms appropriate to a context; used to theorise why AI inference produces a distinct informational injury.
References
- Carayon (1994).Workplace monitoring and electronic performance monitoring.
- Bankins et al. (2024).AI in organisations: efficiency gains and risks to worker experience.
- Foucault (1977).Disciplinary power and visibility.
- Deci and Ryan (2000).Self-determination theory.
- Nissenbaum (2004).Contextual integrity and privacy.
- Brehm (1966).A theory of psychological reactance.
- Hirschman (1970).Exit, Voice, and Loyalty.
- Stanton (2000).Electronic performance monitoring scale.
- Stanton and Julian (2002).Electronic performance monitoring scale.
- Stanton and Weiss (2000).Electronic performance monitoring scale.
- Malhotra et al. (2004).Privacy concern scale.
- Smith et al. (1996).Privacy concern scale.
- Hong and Faedda (1996).Hong reactance scale.
- Dillard and Shen (2005).State operationalisation of reactance.
- Bothma and Roodt (2013).Turnover Intention Scale.
- Hair et al. (2014).A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM).
- Carter (2024).Workplace privacy, people analytics and AI monitoring.
- Jarrahi et al. (2021).Algorithmic management and worker autonomy.
- Ravid et al. (2023).Electronic performance monitoring and employee experience.
- Cardon et al. (2023).Workplace AI and communication privacy.
- Mettler (2023).AI monitoring and employee well-being.
- Yost et al. (2019).Privacy violations and reactance in surveillance contexts.
- Petronio (2015).Communication Privacy Management.
Source & Provenance
sajhrm
Andi Adawiah, Nurmal Idrus
September 14, 2026
Research Study
Asia-Pacific
Original source metadata is preserved. AI analysis is generated separately.
Like this? Get the Monday Decision Brief — free, every week.
No spam, unsubscribe anytime.
Rate this article
Want the full article? Read it at the original source — free, no paywall.
Read original article