Algorithmic management has rapidly emerged as a dominant form of technology-mediated organizational control in the digital economy, reshaping how work is allocated, monitored, evaluated, and coordinated across platforms and traditional firms. This systematic integrative review synthesizes peer-reviewed studies to examine the mechanisms, implications, and tensions of algorithmic control. Drawing on literature from management, information systems, and organizational studies, the review identifies core themes including automated monitoring and surveillance, the automation of managerial functions, worker autonomy and behavioral responses, governance and accountability challenges, and broader effects on organizational design. A novel integrative architecture—the algorithmic management control ecosystem (AMCE) model—is introduced to organize the fragmented research into five interconnected layers. The synthesis reveals persistent tensions between efficiency gains and issues of fairness, transparency, and autonomy, while tracing the evolution of the field from early conceptualizations of big-data-driven control to more recent examinations of platform-specific governance and resistance. Findings highlight how algorithms embed power asymmetries and create new forms of digital Taylorism, yet also open avenues for hybrid human–algorithmic systems. The review concludes by offering a structured foundation for future scholarship on technology-mediated organizational control in digitally transformed workplaces.
The rise of the digital economy has fundamentally altered organizational control mechanisms, shifting authority from human managers to algorithmically embedded systems that automate decision-making, performance tracking, and resource allocation [1-3]. Algorithmic management—defined as the use of data-driven algorithms to coordinate, monitor, and direct work—now permeates gig platforms, traditional corporations, and hybrid work environments alike [4-6]. This transformation is not merely technical; it represents a profound reconfiguration of power relations, managerial roles, and worker experiences within organizations [7-9].
Early research from 2017 onward recognized that big data and advanced analytics enable unprecedented levels of transparency and control, often described as “computer-augmented transparency” that simultaneously augments and supplants traditional hierarchical oversight. By 2019–2020, scholars documented how platform-based labor markets operationalize algorithmic control through real-time rating systems, dynamic pricing, and automated task allocation, creating what some termed a new “contested terrain of control” [10-16]. Subsequent studies (2021–2023) have deepened this analysis by exploring resistance strategies, ethical dilemmas, and the duality of algorithmic systems that both empower and constrain organizational actors [10, 12, 17].
Technology-mediated control manifests in multiple forms: continuous digital surveillance via apps and sensors [3, 8], automated performance evaluation that replaces subjective judgment [2, 6], and predictive algorithms that pre-emptively shape worker behavior [1, 18]. These systems promise efficiency gains and scalability but introduce novel risks, including opacity in decision logic, bias amplification, and erosion of worker autonomy [9, 14, 19]. In gig economies, for instance, algorithms function as de facto managers, dictating work schedules and pay without human intervention, thereby intensifying power asymmetries [4, 16, 20-22].
The implications extend beyond individual workers to organizational design itself. Algorithms reshape structures by flattening hierarchies, enabling remote coordination at scale, and embedding control directly into digital infrastructures [13, 23]. Yet this shift also generates tensions: efficiency versus fairness, control versus autonomy, and transparency versus managerial accountability [11, 12, 15]. Research increasingly questions whether algorithmic governance can be reconciled with ethical and human-centered principles [7, 14, 15].
Despite rapid growth in the literature, the field remains fragmented across disciplines and lacks a unifying synthesis that integrates control mechanisms with behavioral, governance, and design outcomes. Existing reviews tend to focus narrowly on platforms or HRM applications, leaving broader organizational and theoretical implications underexplored [2, 12]. This integrative review addresses that gap by systematically synthesizing peer-reviewed scholarship from 2017 to 2023. It classifies studies into coherent thematic domains, compares contrasting perspectives on algorithmic control, identifies core tensions, and traces the field’s evolution. Central to the analysis is the introduction of an original synthesis model—the Algorithmic Management Control Ecosystem (AMCE)—that organizes the literature into interconnected layers, providing a structured architecture for understanding technology-mediated organizational control.
By mapping how algorithms simultaneously automate managerial functions and reshape human–technology interactions, this review contributes to digital business and management studies. It illuminates how digital infrastructures are not neutral tools but active agents of control that redefine organizational boundaries, roles, and power dynamics in the contemporary economy [1, 3, 11]. The ensuing sections detail the review methodology, present a thematic synthesis, and articulate the AMCE framework, laying the groundwork for a more integrated understanding of algorithmic management.
This systematic integrative review followed a structured, replicable protocol to identify and synthesize peer-reviewed research on algorithmic management and technology-mediated organizational control. The search strategy targeted publications from 2017 to 2023, inclusive, reflecting the period when algorithmic management concepts gained significant traction following advances in platform technologies and big data analytics.
Searches were conducted across major academic databases (Scopus, Web of Science, and Google Scholar) using combinations of keywords and phrases such as “algorithmic management,” “algorithmic control,” “technology-mediated organizational control,” “digital monitoring,” “platform labor management,” “automated managerial functions,” “gig economy algorithms,” and “algorithmic governance.” Results were restricted to English-language, peer-reviewed journal articles and preprints that were subsequently published in peer-reviewed outlets. Targeted journals included Strategic Management Journal, Journal of Business Research, MIS Quarterly, Information & Management, Organization Science, Long Range Planning, Journal of Strategic Information Systems, Technovation, Technological Forecasting & Social Change, Academy of Management Review, and Academy of Management Journal, supplemented by relevant interdisciplinary outlets in information systems, organization studies, and sociology.
Inclusion criteria required that studies (a) explicitly address algorithmic systems as mechanisms of organizational control, (b) examine technology-mediated monitoring, decision automation, or governance in work contexts, and (c) contribute conceptual, empirical, or theoretical insights aligned with the review’s focus. Exclusion criteria eliminated purely technical papers on algorithm design without organizational implications, non-peer-reviewed sources, publications outside the 2017–2023 window, and studies focused solely on consumer-facing algorithms or non-work settings.
After initial retrieval and duplicate removal, titles and abstracts were screened for relevance. Full-text assessment yielded a final corpus of exactly 28 publications that collectively represent the most directly pertinent scholarship. The selected references span diverse methodologies (qualitative case studies, conceptual analyses, and mixed-methods investigations) and contexts (gig platforms, traditional firms, and hybrid digital organizations), ensuring breadth while maintaining thematic coherence. No statistical meta-analysis or PRISMA flow diagram was generated, as the integrative aim was synthesis and theory-building rather than quantitative aggregation. All references were compiled in Vancouver numeric style and are cited consistently throughout the manuscript. This delimited corpus serves as the evidentiary basis for the thematic synthesis and the proposed integrative architecture.
The collected literature reveals a coherent yet multifaceted body of research on algorithmic management that can be synthesized along five emergent thematic domains: algorithmic control and monitoring systems, automation of managerial functions, control and autonomy in digital labor, governance and accountability challenges, and organizational design implications. These domains are not mutually exclusive but intersect, reflecting the systemic nature of technology-mediated control.
Early studies (2017–2019) laid the foundation for these concepts by highlighting how big data infrastructures enable pervasive monitoring and surveillance [8, 13]. Algorithms embedded in platforms capture granular behavioral data—location, task completion, ratings—creating real-time performance dashboards that surpass traditional supervisory methods [3, 16]. This “digital Taylorism” standardizes work while rendering control invisible and automated [5, 22]. By 2020–2021, research shifted toward platform-specific applications, documenting how matching algorithms simultaneously allocate tasks and enforce compliance through rating thresholds and deactivation risks [3, 18, 24]. Workers experience algorithmic management as an omnipresent yet opaque authority that anticipates and shapes behavior before human intervention occurs [1, 10].
A parallel stream examines the automation of managerial roles. Algorithms now perform functions previously reserved for humans: task assignment, performance appraisal, scheduling, and even disciplinary actions [2, 6, 11]. This delegation promises scalability and objectivity but raises questions about the deskilling of managers and the transfer of authority to code [12, 17]. Studies show that algorithmic decision systems reduce managerial discretion while introducing new forms of bias rooted in training data or design choices [9, 19, 25].
The theme of control versus autonomy emerges strongly in gig-economy and app-based work contexts. Workers develop anticipatory compliance strategies—such as gaming ratings or curating digital personas—to navigate algorithmic oversight [4, 5, 8]. Yet research also documents resistance tactics, including collective sensemaking and platform exodus, revealing the contested nature of digital labor [11, 14, 16]. Behavioral adaptations range from heightened self-monitoring to emotional labor aimed at pleasing opaque algorithms [10, 22].
Governance and accountability are growing concerns. Algorithmic opacity complicates responsibility attribution: who is liable when an algorithm unfairly deactivates a worker or miscalculates performance? [7, 14, 15]. Calls for transparency, explainability, and regulatory intervention highlight tensions between proprietary algorithms and public accountability [2, 23, 26]. Ethical dimensions—bias, discrimination, and fairness—further complicate governance, as algorithms can amplify existing inequalities [9, 19].
Finally, organizational design implications are addressed in later scholarship. Algorithmic control flattens hierarchies, enables boundaryless coordination, and reconfigures power distribution across digital infrastructures [12, 13, 18]. However, it also creates hybrid structures where human oversight coexists uneasily with automated systems, prompting new questions about organizational resilience and adaptation [23, 27-29].
Over the 2017–2023 period, the literature shifts from descriptive accounts of control mechanisms to critical analyses of power, ethics, and resistance. Comparative perspectives reveal consensus on efficiency gains alongside divergence over normative outcomes: some emphasize the liberatory potential of data-driven optimization [2, 13], while others foreground exploitation and dehumanization [4, 7, 16]. Persistent tensions—efficiency versus fairness, control versus autonomy, opacity versus accountability—run through all domains, underscoring the need for integrative frameworks. The following section organizes these insights into a unified architecture. Table 1 clarifies that the central contradictions in algorithmic management are best understood as cross-layer tensions within the AMCE ecosystem, rather than as isolated problems of monitoring, automation, or governance alone.
Table 1. Cross-layer tensions in the AMCE
Tension axis | Primary AMCE layers involved | What produces the tension | Organizational manifestation | Likely consequences | Key analytical implication |
Efficiency ↔ Fairness | Algorithmic control and monitoring; managerial automation and decision; governance, transparency, and accountability | Data-driven optimization privileges speed, scale, and standardization, while fairness requires contextual interpretation, explainability, and procedural safeguards | Rapid task allocation, automated evaluation, deactivation thresholds, and standardized performance scoring | Higher throughput and lower coordination costs, but increased risk of bias, procedural opacity, and contested legitimacy | Efficiency gains in algorithmic management are not neutral outcomes; they are achieved through control logics that often redistribute risk downward to workers and governance structures |
Control ↔ Autonomy | Algorithmic control and monitoring; managerial automation and decision; worker autonomy and behavioral adaptation | Continuous monitoring and predictive coordination narrow discretion, while workers seek room for adaptation, sensemaking, and resistance | Anticipatory compliance, rating management, schedule gaming, selective disengagement, and platform exit | Short-term compliance but long-term strain, emotional labor, defensive behavior, and resistance cycles | Worker agency is not eliminated by algorithmic control; it is reconfigured into adaptive, tactical, and sometimes collective responses |
Opacity ↔ Accountability | Managerial automation and decision; governance, transparency, and accountability; organizational design and structural outcomes | Decision rules are embedded in proprietary code and data models, while accountability requires traceability, explainability, and assignable responsibility | Unclear basis for sanctions, weak appeal mechanisms, diffusion of responsibility between managers, designers, and platforms | Erosion of trust, legal exposure, reputational risk, and governance gaps in hybrid organizations | The more decision authority is embedded in algorithms, the more organizational accountability must be redesigned rather than assumed |
Standardization ↔ Context sensitivity | Algorithmic control and monitoring; managerial automation and decision; worker autonomy and behavioral adaptation | Scalable systems depend on uniform metrics, but work situations vary by task, customer interaction, and local conditions | Universal metrics applied to heterogeneous jobs, with limited recognition of situational constraints | Misclassification of performance, perceived injustice, and reduced worker voice | Algorithmic control systems are strongest where tasks are highly codifiable and weakest where context materially affects performance |
Scalability ↔ Human oversight | Managerial automation and decision; governance, transparency, and accountability; organizational design and structural outcomes | Automation expands managerial reach, but consequential decisions still require review, judgment, and exception handling | Lean supervision, dashboard management, and escalation only for edge cases | Lower labor costs, but weakened managerial learning and reduced corrective capacity | Hybrid oversight is not an optional ethical add-on; it is an organizational design requirement for sustainable algorithmic governance |
Behavioral optimization ↔ Human dignity | Algorithmic control and monitoring; worker autonomy and behavioral adaptation; governance, transparency, and accountability | Systems seek measurable behavioral conformity, while workers seek recognition beyond quantifiable outputs | Self-surveillance, emotional regulation, persona curation, and algorithm-facing labor | Alienation, perceived dehumanization, and performative compliance | Digital control extends beyond task execution into identity and affect, making dignity a governance issue rather than merely a cultural one |
To synthesize the fragmented yet rapidly expanding literature, this review introduces the AMCE model. The AMCE provides a layered, non-causal architecture that organizes reviewed studies into five interdependent dimensions, illustrating how algorithmic control operates as an interconnected system rather than isolated mechanisms.
The five thematic dimensions are:
Algorithmic control and monitoring layer: Encompasses data capture, surveillance, and real-time performance tracking systems that embed control within digital infrastructures [1, 3, 8, 13, 16, 22, 24].
Managerial automation and decision layer: Captures the delegation of core managerial functions (allocation, evaluation, coordination) to algorithms [2, 6, 11, 12, 17, 19, 25].
Worker autonomy and behavioral adaptation layer: Addresses how individuals and collectives respond to, resist, or comply with algorithmic directives, shaping emergent behaviors [4, 5, 10, 14, 18, 28].
governance, transparency, and accountability layer: Examines regulatory, ethical, and responsibility structures surrounding opaque algorithmic systems [7, 9, 15, 23, 26].
Organizational design and structural outcomes layer: Explores resulting changes in hierarchy, boundaries, and power distribution within digitally transformed organizations [12, 13, 18, 23].
A review-synthesis diagram of the AMCE framework is shown in Figure 1.
Figure 1. A review-synthesis diagram of the AMCE framework
This AMCE architecture integrates the reviewed literature by positioning algorithmic control as an ecosystem rather than a linear process. It highlights feedback loops whereby behavioral adaptations influence monitoring design, and organizational outcomes recursively shape governance requirements [10, 12, 14]. By mapping the corpus onto these layers, the model reveals both convergence (e.g., a universal emphasis on monitoring efficiency) and divergence (e.g., between optimistic and critical views of autonomy). The framework thus serves as an organizing lens for future research while underscoring the systemic character of technology-mediated organizational control in the digital economy.
The AMCE framework reveals that algorithmic management operates as a tightly coupled ecosystem in which each layer dynamically influences the others, generating persistent tensions that have intensified across the 2017–2023 period [1, 11, 14]. The Algorithmic Control and Monitoring Layer does not function in isolation; its surveillance mechanisms feed directly into the Managerial Automation and Decision Layer, enabling predictive scheduling and automated sanctions that reshape worker behavior in real time [3, 22, 24]. This feedback produces anticipatory compliance documented across gig-platform studies, where rating algorithms compel workers to internalize control logics before any human manager intervenes [5, 10, 18].
Simultaneously, the worker autonomy and behavioral adaptation layer generates counter-loops: resistance practices and collective sensemaking occasionally force recalibration of monitoring algorithms, illustrating the contested terrain of control [11, 14, 16]. Governance, Transparency, and Accountability Layer tensions become most visible when opacity in the Managerial Automation Layer collides with demands for fairness, as biased performance metrics amplify inequalities that, in turn, constrain Organizational Design and Structural Outcomes [7, 9, 15, 19, 25]. A comparative analysis of the corpus shows an evolution from early optimism about efficiency gains through computer-augmented transparency [13] to later recognition of systemic power asymmetries and digital Taylorism [4, 22, 28]. These cross-layer dynamics underscore that algorithmic control is neither deterministic nor neutral; it emerges from recursive interactions between technology, human actors, and organizational structures [2, 12, 23].
This integrative review advances theory by offering the AMCE Model as a unifying architecture that transcends discipline-specific silos. Prior conceptual work treated algorithmic management primarily as an HRM or platform phenomenon [6, 17]; the AMCE reframes it as a multi-layered ecosystem that integrates control theory, socio-technical systems perspectives, and organizational design scholarship [1, 11, 13, 23]. By classifying the studies into five interdependent dimensions and explicitly mapping feedback loops and tension arrows, the model provides a non-linear lens for future theorizing. This architectural contribution is significant because it acknowledges that algorithmic management is not a monolithic construct but rather a complex system in which technological infrastructures, governance mechanisms, behavioral responses, and managerial interventions continuously interact and co-evolve. The model moves beyond static depictions of “algorithmic control” toward a dynamic ecosystem view that accounts for mutual reinforcement between monitoring infrastructures and behavioral responses [10, 14]. In doing so, it recognizes that control in technology-mediated environments is not simply imposed from above but emerges through recursive interactions: monitoring shapes behavior, behavioral adaptations inform system refinements, and these refinements in turn generate new forms of resistance or accommodation.
Table 2 consolidates the manuscript’s dispersed theoretical foundations by showing that no single lens fully explains algorithmic management, thereby justifying the AMCE model as a multi-perspective integrative architecture.
Table 2. Theoretical consolidation matrix for research on algorithmic management
Theoretical lens | Core analytical question | What it explains well in algorithmic management | What it tends to underexplain | Main AMCE layers illuminated | Value added for future research |
Control theory | How are behavior and outputs directed toward organizational goals? | Formalization of monitoring, performance metrics, compliance mechanisms, and substitution of hierarchical supervision with digital controls | The political and ethical implications of who designs the control architecture and who bears its risks | Algorithmic control and monitoring; managerial automation and decision | Enables sharper distinctions between input, process, and output control under algorithmic conditions |
Labor process theory/digital Taylorism | How does technology reorganize labor extraction, discipline, and power asymmetry? | Intensification, deskilling, surveillance, the invisibility of managerial authority, and worker subordination through code | Situations where workers retain meaningful discretion or where algorithms augment rather than simply dominate labor | Algorithmic control and monitoring; worker autonomy and behavioral adaptation; organizational design and structural outcomes | Helps explain why efficiency gains are often experienced as domination rather than empowerment |
Sociotechnical systems theory | How do technical infrastructures and human actors co-produce organizational outcomes? | Mutual shaping of algorithms, workers, managers, and routines; recursive adaptation and hybrid arrangements | Macro-political issues such as regulation, ownership, and structural inequality | All five AMCE layers | Particularly useful for testing feedback loops and non-linear ecosystem dynamics |
Organizational design theory | How do technologies reshape hierarchy, coordination, boundaries, and authority structures? | Flattening of middle management, distributed coordination, new boundary configurations, platform-centric structuring | Worker-level experience, micro-resistance, and fairness perceptions | Managerial automation and decision; organizational design and structural outcomes | Clarifies how algorithmic management changes the architecture of the firm, not only supervision practices |
Governance and accountability theory | Who is responsible for consequential decisions made through opaque sociotechnical systems? | Explainability, contestability, liability, procedural justice, and regulatory oversight | Everyday adaptation at the point of work and informal coping practices | Governance, transparency, and accountability; managerial automation and decision | Provides a foundation for studying appeals, auditability, and institutional responses across sectors |
Behavioral adaptation/sensemaking perspectives | How do workers interpret, comply with, game, or resist algorithmic control? | Anticipatory compliance, emotional labor, tactical gaming, collective interpretation, resistance, and exit | System-level design logics and broader organizational restructuring | Worker autonomy and behavioral adaptation | Helps move beyond deterministic views by foregrounding agency under constrained conditions |
Sociomateriality | How are control, agency, and organizational action constituted through the entanglement of technology and practice? | Non-separability of algorithmic systems and organizational routines; distributed agency among managers, workers, and code | Normative questions about fairness and regulation, unless combined with governance perspectives | All five AMCE layers, especially interdependencies | Supports the manuscript’s claim that algorithms are not neutral tools but active organizational agents |
Ethical AI/responsible innovation | How should algorithmic systems be designed and governed to protect fairness and human dignity? | Bias, transparency, accountability, explainability, and human oversight principles | The everyday realities of labor control and contested compliance in organizational settings | Governance, Transparency and Accountability; Managerial Automation and Decision | Offers normative design criteria that can be integrated with management theory for more actionable research |
Theoretically, the AMCE highlights how technology-mediated control reconstitutes power without eliminating human agency, thereby bridging critical management studies with information-systems research on sociomateriality [3, 18, 12]. This bridging is essential because prior literature has often talked past one another: critical perspectives have emphasized surveillance and power asymmetries. At the same time, socio-technical scholarship has focused on system design and implementation. The AMCE integrates these viewpoints by demonstrating that agency is not erased but redistributed—managers, workers, and algorithms each exercise forms of influence that are simultaneously constrained and enabled by technological architectures. This contribution fills a noted gap in the literature for holistic frameworks capable of synthesizing the rapid proliferation of empirical findings between and [2, 26]. The timeframe in question witnessed an explosion of empirical studies as algorithmic management moved from a niche concern in platform work to a mainstream organizational phenomenon spanning logistics, retail, professional services, and manufacturing. Without a unifying architecture, these findings risked remaining fragmented; the AMCE provides the integrative structure necessary to identify patterns, contradictions, and boundary conditions across contexts.
Managers and platform operators face concrete imperatives derived from the AMCE layers. First, organizations must design hybrid oversight mechanisms that retain human judgment within the Managerial Automation Layer to mitigate bias amplification and restore accountability [9, 19, 25]. Hybrid oversight functions as a safeguard against the tendency of algorithmic systems to perpetuate or even magnify existing biases embedded in historical data, while also ensuring that decisions with significant consequences for workers are subject to meaningful human review rather than fully automated adjudication. This requires not merely occasional human intervention but systematically designed touchpoints where managerial discretion serves as a corrective to algorithmic outputs.
Second, transparency-enhancing interventions—such as explainable rating dashboards or worker-accessible algorithm audits—can reduce opacity tensions and foster trust within the Governance Layer [7, 15, 23]. Opacity has been consistently identified as a source of worker anxiety, perceived unfairness, and resistance; conversely, when workers understand how decisions affecting them are made, they report greater acceptance even when outcomes are unfavorable. Transparency mechanisms must be substantive rather than symbolic, providing genuine insight into system logic rather than superficial disclosure that obscures underlying complexity.
Third, recognizing behavioral adaptation loops, firms should incorporate worker voice mechanisms (e.g., collective feedback channels) to prevent destructive resistance and harness autonomy constructively [5, 10, 14]. Behavioral adaptation loops capture the reality that workers subjected to algorithmic management are not passive recipients but active agents who learn, adapt, and often develop counter-strategies—from gaming rating systems to collectively challenging opaque decisions. Formalizing worker voice transforms this energy from a source of friction into a source of system improvement, enabling organizations to detect unintended consequences early and refine algorithmic governance accordingly.
Finally, organizational redesign efforts should treat algorithmic infrastructures as strategic assets that simultaneously flatten hierarchies and redistribute power, requiring proactive governance policies rather than reactive compliance [12, 13, 18]. Organizations that approach algorithmic management reactively—adopting technologies without considering their governance implications—often find themselves confronting unanticipated consequences ranging from erosion of worker trust to regulatory exposure. Proactive governance, by contrast, embeds considerations of fairness, accountability, and autonomy into system design from the outset, anticipating tensions rather than responding to crises. These practices enable firms to capture efficiency gains while mitigating fairness and autonomy trade-offs inherent in technology-mediated control.
The review is limited to peer-reviewed sources that explicitly address algorithmic systems as organizational control mechanisms. Consequently, it excludes pre- foundational work, purely technical algorithm-development papers, and studies focused exclusively on consumer or non-work algorithms. The decision to anchor the review in this timeframe reflects the concentration of empirical organizational research on algorithmic management within these years; prior work was predominantly conceptual or focused on adjacent topics such as electronic performance monitoring. While the corpus spans leading journals in management and information systems, it does not encompass every interdisciplinary outlet; grey literature and non-English publications were omitted to maintain peer-review rigor. This deliberate delimitation ensures that all synthesized findings have undergone rigorous editorial review and contribute to a coherent evidentiary base. However, it also means that insights from fields such as human-computer interaction, labor studies, and critical data studies may be underrepresented. The synthesis prioritizes conceptual integration over quantitative meta-analysis, limiting generalizability of effect sizes but enhancing depth of thematic interconnection. These boundaries reflect the deliberate focus on emerging research trajectories within the specified time frame, capturing a period of rapid empirical growth while maintaining methodological discipline.
Future research should empirically test the AMCE layers through longitudinal designs that track feedback loops across platform and traditional organizational contexts [1, 12]. Cross-sectional studies have established correlations; longitudinal research is now needed to understand how algorithmic management systems evolve, how feedback loops manifest across different temporal scales, and whether initial effects persist or attenuate as actors adapt. Comparative studies examining cultural and regulatory variations in governance responses would enrich the Accountability Layer [15, 26]. Algorithmic management unfolds within diverse institutional environments; understanding how cultural norms around fairness, regulatory frameworks for algorithmic accountability, and collective bargaining structures moderate system effects would substantially extend the model’s explanatory power.
Interdisciplinary collaborations integrating ethical AI frameworks with organizational control theory are needed to address fairness tensions [9, 19]. Ethical AI scholarship has developed sophisticated frameworks for fairness, accountability, and transparency in algorithmic systems, yet these have rarely been systematically integrated with organizational theory on control, authority, and resistance. Bridging these fields would yield both theoretically robust and practically actionable insights. Longitudinal field experiments on hybrid human–algorithmic decision systems could illuminate conditions under which automation enhances rather than erodes managerial roles [2, 6, 17]. Such experiments would help move beyond correlational evidence toward causal understanding, identifying the specific design features, implementation processes, and contextual factors that shape whether algorithmic management augments human capability or substitutes for managerial judgment.
Finally, scholars should explore post- developments in generative AI and edge-computing surveillance to update the ecosystem model for evolving technological realities. The period following the review’s temporal boundary has witnessed rapid advances in generative AI capabilities and the proliferation of surveillance technologies operating at the edge of organizational networks. These developments introduce new governance challenges—including questions of synthetic decision-making, real-time behavioral modification, and the blurring of work and non-work boundaries—that existing frameworks are only beginning to address. Updating the AMCE model to account for these emerging technologies represents a critical frontier for sustaining its relevance and explanatory power.
Algorithmic management has matured into a pervasive form of technology-mediated organizational control that redefines managerial authority, worker experience, and organizational architecture in the digital economy. By synthesizing 28 key studies using the AMCE framework, this review demonstrates that control, automation, autonomy, governance, and structural outcomes form an interdependent ecosystem characterized by inescapable tensions between efficiency and fairness, and between opacity and accountability. The AMCE Model offers scholars and practitioners a coherent architecture for navigating these dynamics, underscoring that algorithms are not merely tools but active agents that both constrain and enable new organizational possibilities. As digital infrastructures continue to evolve, sustained critical inquiry into technology-mediated control remains essential for fostering workplaces that balance productivity with human dignity.
None
None
None
None
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Institute for Management, Business, and Accounting Studies
