The integration of artificial intelligence (AI) into strategic management has transformed how organizations design decision systems and pursue competitive advantage in digital environments. This narrative literature review synthesizes peer-reviewed studies, focusing on AI-enabled strategic decision-making, algorithmic organizational systems, and the mechanisms through which AI generates sustained performance gains. Drawing on top-tier journals such as Strategic Management Journal, MIS Quarterly, and Technological Forecasting and Social Change, the analysis identifies recurring patterns of augmentation rather than replacement. It surfaces persistent tensions in human–AI collaboration and governance. Key findings show that AI augments predictive accuracy and resource allocation, yet introduces novel risks related to algorithmic bias, ethical oversight, and the erosion of traditional sources of advantage. The review traces the field’s evolution from early conceptual explorations of human–AI symbiosis to more recent examinations of generative AI’s disruptive potential and firm-level outcomes. Conceptual overlaps emerge around the centrality of hybrid decision architectures, while inconsistencies appear in assessments of long-term competitive sustainability. By mapping these streams and their interrelationships, the manuscript offers a structured foundation for understanding AI’s strategic role. It highlights critical gaps in cross-industry generalizability, ethical frameworks, and the interplay between technological affordances and organizational adaptation. This synthesis equips scholars and executives with an integrated lens on how AI is reconfiguring strategic management in the digital age.
Artificial intelligence has moved from a peripheral technological curiosity to a core strategic asset [1], fundamentally altering how firms sense opportunities, allocate resources, and execute strategy [2-5]. Since 2017, the accelerating capabilities of machine learning, generative models, and predictive analytics have prompted organizations to embed AI directly into decision architectures [2], shifting strategic processes from intuition-driven hierarchies to data-augmented, algorithmically supported systems [6, 7]. This shift is not merely technical; it reshapes the very sources of competitive advantage by enabling faster environmental scanning, superior pattern recognition, and more precise resource orchestration.
Early conceptual work emphasized AI’s potential to augment rather than replace managerial judgment. Subsequent empirical and theoretical contributions have tested these ideas across industries [3], revealing both performance uplifts and unanticipated organizational frictions. For instance, studies in MIS Quarterly demonstrate that CIOs and boards play pivotal roles in aligning AI initiatives with strategic priorities [6]. At the same time, hybrid human–AI systems improve operational outcomes when properly configured [2, 3].
Yet the literature also documents significant challenges. Algorithmic decision-making can amplify biases embedded in training data [8-10], erode employee autonomy, or create over-reliance on opaque systems. Competitive advantage derived from AI is increasingly viewed as situated and context-dependent rather than universally sustainable, with imitation and rapid diffusion threatening early-mover gains. Moreover, the rise of generative AI has introduced new questions about creative problem-solving and employment dynamics [11], further complicating strategic workforce planning [12, 13].
This narrative review addresses these developments by synthesizing research at the intersection of AI, organizational decision systems, and competitive advantage. Unlike systematic reviews that prioritize statistical aggregation, this interpretive synthesis compares theoretical streams, highlights conceptual overlaps and disagreements, and traces the field’s maturation from 2017 conceptual optimism to 2024 evidence-based caution. The analysis draws exclusively on peer-reviewed work from leading management and information systems journals to ensure scholarly rigor and relevance.
To produce a focused yet comprehensive narrative review, a targeted yet interpretive search protocol was implemented across major academic databases (Scopus, Web of Science, Google Scholar, and journal-specific portals). The temporal window was deliberately restricted to 2017–2024 inclusive [2], capturing the period of exponential AI advancement following the deep-learning breakthrough and preceding the widespread commercialization of generative models [7]. Search strings combined core terms (“artificial intelligence” OR “machine learning” OR “algorithmic decision-making”) with strategic management constructs (“strategic decision” OR “competitive advantage” OR “organizational decision systems” OR “firm performance” OR “digital transformation”). Additional filters targeted the specified high-impact journals: Strategic Management Journal, Journal of Business Research, MIS Quarterly, Information & Management, Organization Science, Long Range Planning, Journal of Strategic Information Systems, Technovation, and Technological Forecasting & Social Change.
Initial results exceeded 1,200 records. Iterative screening applied strict inclusion criteria: (1) peer-reviewed journal articles (preprints retained only if subsequently published in qualifying outlets); (2) explicit focus on AI’s role in strategic processes, decision systems, or competitive outcomes [3]; (3) empirical, conceptual, or review contributions offering theoretical insight rather than purely technical descriptions [4]; and (4) direct relevance to organizational contexts [6]. Exclusion removed purely technical AI papers, non-management disciplines, and studies outside the date range [10]. Forward and backward citation tracking supplemented the database searches [11], ensuring saturation of influential streams [13].
This narrative approach—rather than a strictly systematic meta-analysis—was chosen to permit interpretive synthesis of conceptual patterns [1], identification of theoretical tensions [2], and tracing of temporal evolution [5]. The selected works span multiple methodologies (qualitative case studies, survey-based analyses, conceptual theorizing, and econometric examinations) and geographic contexts, enhancing robustness [7]. Journal diversity ensures balance between strategy, information systems, and innovation perspectives [14-20]. All citations in the manuscript adhere strictly to the compiled reference list, preserving traceability and fidelity to the source material [21, 22]. The resulting synthesis therefore reflects the current state of knowledge while remaining transparent about its boundaries and deliberate focus on strategic management implications [23-28].
Table 1 clarifies how the five thematic clusters differ in their analytical focus, strategic mechanisms, and unresolved tensions, thereby revealing the field’s internal structure rather than treating the literature as a single undifferentiated stream.
Table 1. Analytical distinctions across the five thematic clusters in AI and strategic management research
Thematic cluster | Primary analytical focus | Dominant unit of analysis | Central mechanism | Strategic contribution | Principal unresolved tension |
AI-enabled strategic decision systems [2, 3, 6, 7, 22, 26] | Design of hybrid decision architectures combining algorithmic analytics and managerial judgment | Decision process/organizational system | AI processes high-volume data while humans provide contextual interpretation and final calibration | Improves speed, consistency, and allocative precision in strategic decision-making | How much authority can be delegated to AI before strategic misalignment or overdependence emerges? |
Competitive advantage through AI capabilities [4, 24, 25, 27, 28] | How AI resources and capabilities generate firm differentiation | Firm/capability configuration | Data, models, and analytics are combined with organizational complements to create situated advantage | Reframes competitive advantage as data-intensive, learning-dependent, and context-specific | Whether an AI-based advantage can remain durable under imitation, diffusion, and commoditization pressures |
Human–AI collaboration in strategy [11-13, 22] | Interaction between human expertise and algorithmic outputs | Individual/team/socio-technical interface | Distributed cognition and calibrated hybridity improve problem solving, interpretation, and adaptability | Preserves creativity, judgment, and resilience while exploiting AI efficiency | Where augmentation ends, and substitution begins, especially in generative AI contexts |
Algorithmic decision-making, control, and ethical governance [10, 23] | Oversight, accountability, transparency, and lifecycle control of AI systems | Governance structure/organizational control system | Monitoring, auditability, and ethical review shape whether AI systems remain legitimate and governable | Protects trust, legitimacy, and long-term viability of AI-enabled decision systems | How governance can remain effective when models are opaque, adaptive, and rapidly scaled |
AI-driven digital transformation and organizational renewal [4, 5, 14-19] | Organization-wide capability reconfiguration through AI adoption | Organization/ecosystem/transformation process | AI is embedded into innovation, value-chain redesign, and capability renewal | Connects AI adoption to broader strategic renewal and business model adaptation | Why do some firms convert AI adoption into transformation, while others experience misfit and fragmentation |
The literature coalesces around five tightly interrelated research themes that, together, explain how artificial intelligence is reconfiguring strategic management [1]. These themes are not isolated silos but form a dynamic system in which technologies shape decision architectures [2], which in turn influence competitive positioning [3], moderated by human and governance factors [5].
A foundational stream examines the architecture of hybrid decision systems that blend algorithmic prediction with human judgment [2, 3, 6, 7, 22, 26]. Trunk et al. [2] demonstrate that organizations achieve superior outcomes when AI handles data-intensive analytics while executives retain contextual oversight. Al-Surmi et al. [3] provide evidence that AI-driven decision models enhance operational performance across supply-chain and market-entry contexts. MIS Quarterly contributions further underscore the strategic importance of CIO and board-level governance in directing these systems [6, 7]. The theme reveals a consistent shift from static, hierarchical decision processes to dynamic, learning-oriented architectures capable of real-time adaptation.
A second cluster investigates the mechanisms by which AI generates and sustains advantage [4, 24, 25, 27, 28]. Kemp [24] develops a situated theory of AI advantage emphasizing context-specific deployment. Krakowski et al. [28] document how AI alters the traditional resource-based view by making data and algorithmic capabilities primary sources of differentiation. Tong et al. [27] reveal Janus-faced effects whereby AI feedback simultaneously boosts and constrains employee performance. Overlaps with Theme 1 are evident: robust decision systems become the vehicle for realizing advantage, yet the literature cautions that advantages may prove temporary without continuous organizational learning [28].
Research in this theme emphasizes symbiotic rather than substitutive relationships [12, 13, 22]. Anthony et al. [12] advocate a systems perspective on future work arrangements, while Jarrahi [22] conceptualizes decision-making as distributed cognition between humans and algorithms. Recent Organization Science studies highlight the short-term employment effects of generative AI [13]. The theme surfaces productive tensions: collaboration enhances creativity and resilience, yet requires deliberate design to prevent deskilling or over-delegation.
Algorithmic systems introduce new challenges in control and accountability [10, 23]. Marabelli et al. [23] trace the lifecycle of such systems, identifying ethical inflection points at design, deployment, and monitoring stages. Editorials in MIS Quarterly call for greater scholarly attention to transparency and bias mitigation [10]. This theme intersects with all others, underscoring that without appropriate governance, even technically superior decision systems can undermine long-term competitive legitimacy.
The final cluster situates AI within broader transformation processes [4, 5, 14, 15]. Multiple Technological Forecasting and Social Change studies illustrate how AI accelerates innovation, reshapes R&D, and reconfigures value chains [15-17]. Sestino and De Mauro [5] map practical applications and methods, while Perifanis and Kitsios [4] link AI adoption to measurable business value. The theme emphasizes that competitive advantage emerges not from technology alone but from its orchestration within transformed organizational capabilities. Figure 1 illustrates the interdependent architecture through which AI capability inputs are translated into competitive outcomes via hybrid organizational decision systems shaped by collaboration, governance, and digital transformation.
Figure 1. Interdependent architecture of AI in strategic management: how decision systems translate AI capabilities into competitive advantage. The figure synthesizes the manuscript’s core theoretical argument that artificial intelligence influences firm-level competitive outcomes through AI-powered organizational decision systems rather than through technology adoption alone. AI capability inputs feed a central hybrid decision architecture that interacts recursively with four interdependent domains: strategic decision augmentation, competitive advantage formation, human–AI collaboration, and governance and ethical control. These relationships are embedded within broader processes of AI-driven digital transformation and organizational renewal. Competitive outcomes emerge conditionally from this system and are continually reshaped through feedback, learning, and governance adjustment. The dashed outer horizon highlights unresolved research questions regarding sustainability, legitimacy, generalizability, and the long-term evolution of AI-enabled strategic capabilities.
Building directly on the five thematic clusters identified earlier, this section deepens the analysis by comparing research streams, surfacing explicit overlaps, and exposing theoretical frictions that have shaped the field.
Early contributions framed decision systems as static hybrids [2, 22], yet later work demonstrates dynamic, learning-oriented architectures that adapt in real time [3, 6, 7]. Trunk et al. [2] and Al-Surmi et al. [3] converge on performance gains when AI manages data volume while humans supply contextual nuance. Yet, MIS Quarterly studies [6, 7] introduce a critical layer: without CIO and board oversight, these systems risk strategic misalignment. The overlap with Theme 5 is pronounced—digital transformation accelerates only when decision architectures are re-engineered rather than merely automated [4, 17].
Kemp’s situated theory [24] and Krakowski et al.’s reconfiguration of the resource-based view [28] overlap in asserting that advantage derives from context-specific deployment rather than raw technology ownership. However, Tong et al. [27] expose the Janus-faced reality: AI feedback boosts short-term performance but can constrain long-term creativity. This tension directly challenges the optimism of earlier capability-focused studies [25], illustrating that competitive edges erode rapidly without continuous organizational learning [28].
Anthony et al. [12] and Jarrahi [22] converge on distributed cognition as the optimal model, yet Organization Science evidence on generative AI [11, 13] reveals substitution pressures in creative and routine tasks alike. The literature, therefore, moves beyond simple augmentation narratives toward calibrated symbiosis—human oversight remains indispensable precisely because algorithmic outputs require interpretive translation [2, 26].
Marabelli et al. [23] and MIS Quarterly editorials [10] intersect with every preceding theme, demonstrating that governance is not an add-on but the linchpin of sustainable systems. Without lifecycle oversight, even technically superior decision architectures undermine legitimacy [23], a point echoed across innovation studies [16, 19].
Technological forecasting and social change contributions [14-19] collectively trace how AI reshapes R&D, value chains, and innovation processes, yet integration success hinges on the decision-system foundations established in Theme 1 [4, 5]. The synthesis exposes a clear evolutionary trajectory. Papers emphasized opportunity [21, 22], while 2021–2024 research foregrounds friction, contingency, and governance [23, 24, 28].
Across the reference set, a consistent pattern emerges: augmentation-centric configurations tend to outperform full automation in terms of productivity, adaptability, and decision quality [2, 3, 12, 22]. Yet, this empirical regularity is increasingly destabilized by evidence from generative AI contexts, where substitution effects are accelerating—even in domains historically considered resistant to automation, such as creative and knowledge-intensive work [11, 13]. This apparent contradiction reveals a deeper structural tension rather than a simple empirical inconsistency.
Specifically, augmentation and automation should not be conceptualized as mutually exclusive endpoints on a linear continuum, but as dynamically interdependent modes of organizing work. Augmentation enhances human judgment under conditions of uncertainty and equivocality, whereas automation excels in environments characterized by codifiability and scale. Generative AI blurs this boundary by extending automation into semi-structured and symbolic domains, thereby compressing the space in which augmentation traditionally dominated.
The strategic implication is that competitive advantage increasingly hinges on calibrated hybridity: the organizational capacity to continuously reconfigure the division of labor between humans and algorithms. Firms that over-rotate toward full automation risk brittleness, loss of tacit knowledge, and diminished problem-framing capabilities. Conversely, those that rely excessively on augmentation may forgo efficiency gains and scalability. The frontier, therefore, lies in designing adaptive socio-technical systems that dynamically allocate tasks based on contextual complexity, rather than static technological preferences.
The durability of AI-driven advantage remains highly contested. While early work emphasized the potential for data accumulation and algorithmic sophistication to generate sustained competitive advantages [25], more recent contributions converge on a markedly more pessimistic assessment. Kemp [24] and Krakowski et al. [28] argue that AI-based advantages are inherently situated, transient, and vulnerable to rapid erosion.
This shift reflects the declining strength of traditional isolating mechanisms. Data, once presumed to be a defensible asset, is increasingly replicable or accessible through platform ecosystems, synthetic generation, or data-sharing arrangements. Similarly, algorithmic innovations diffuse quickly due to open-source communities, cloud-based AI services, and standardized tooling. As a result, the locus of advantage migrates away from technical artifacts toward organizational complements—including routines, decision processes, and integration capabilities [27, 28].
From a resource-based perspective, this suggests that AI, in isolation, rarely satisfies the VRIN (valuable, rare, inimitable, non-substitutable) criteria. Instead, advantage emerges from the co-specialization between AI systems and firm-specific organizational processes. This reframing aligns with dynamic capability theory: what matters is not possession of AI assets per se, but the ability to orchestrate, recombine, and redeploy them in response to shifting environments. Consequently, sustainability is less about technological superiority and more about organizational agility and learning velocity.
Algorithmic opacity, bias, and accountability risks fundamentally reshape the strategic role of governance. Rather than functioning as a peripheral ethical concern, governance emerges as a central capability that directly influences firm legitimacy, stakeholder trust, and long-term performance [10, 23].
The literature underscores that AI systems introduce novel forms of epistemic risk: decision processes become less interpretable, causal attributions more ambiguous, and responsibility more diffuse. In such contexts, failures are not merely operational but reputational and institutional. Firms that neglect lifecycle accountability—spanning data sourcing, model development, deployment, and monitoring—face escalating regulatory scrutiny and erosion of stakeholder confidence [23, 24].
Strategically, governance must therefore be reconceptualized as an enabling infrastructure rather than a constraint. Effective governance architectures integrate technical mechanisms (e.g., auditability, explainability), organizational processes (e.g., cross-functional oversight, escalation protocols), and institutional alignment (e.g., compliance with evolving regulatory regimes). Firms that internalize governance as a core capability are better positioned to convert trust into a competitive asset, particularly in high-stakes domains where legitimacy is a prerequisite for market participation.
AI-driven digital transformation simultaneously amplifies opportunities for innovation and exposes organizations to new forms of vulnerability. On one hand, AI accelerates recombination, experimentation, and business model renewal, enabling firms to sense and seize opportunities with unprecedented speed [15-17]. On the other hand, these same technologies intensify competitive pressures and magnify the consequences of organizational misalignment.
A key insight across the literature is that technological adoption alone does not guarantee positive outcomes. Instead, performance effects depend on the alignment between AI systems and decision architectures—the structures through which information is processed, interpreted, and acted upon [4, 13, 19]. When decision rights, incentives, and human capabilities lag behind technological deployment, organizations experience what might be termed “algorithmic misfit,” leading to degraded decision quality and strategic incoherence.
This duality reframes digital transformation as a fundamentally contingent process. Success depends less on the extent of AI adoption and more on the coherence between technology, organizational design, and human capital. Firms that treat AI as an isolated technological upgrade risk disruption; those that embed it within a broader transformation of decision-making and capability development are more likely to realize sustained benefits.
Despite the field’s rapid maturation, significant gaps remain. A dominant limitation is the scarcity of longitudinal evidence. Most studies provide cross-sectional snapshots, offering limited insight into how AI capabilities evolve, decay, or recombine over time [3, 28]. This constrains our ability to theorize about path dependence, learning curves, and the long-term sustainability of AI-enabled advantages.
Moreover, the empirical base remains skewed toward large, resource-rich organizations in advanced economies. Small and medium-sized enterprises (SMEs) and non-Western contexts are underrepresented, raising concerns about external validity and limiting our understanding of how resource constraints, institutional environments, and cultural factors shape AI adoption and outcomes [4, 5].
Ethical and governance frameworks, particularly in the context of generative AI, are still embryonic. While concerns around bias, transparency, and accountability are well documented [11, 23], their integration into strategic decision-making remains under-theorized. Equally notable is the near absence of research on the intersection between AI and non-market strategy, including regulatory engagement, political risk management, and stakeholder activism.
Looking forward, several promising avenues emerge. First, comparative studies examining hybrid versus fully algorithmic systems can clarify boundary conditions under which each configuration excels. Second, micro-foundational research on managerial cognition and judgment in AI-augmented environments can illuminate how decision-makers interpret and rely on algorithmic outputs. Third, longitudinal designs are essential to assess the durability of competitive advantages in the wake of the post-2023 wave of generative AI.
Finally, advancing theory in this domain requires deeper integration with adjacent fields, including behavioral strategy, information systems ethics, and innovation policy. Such cross-disciplinary synthesis is critical for developing multi-level frameworks that capture the complex interplay between technology, organizations, and institutions. Only through this broader lens can scholarship fully account for the strategic ramifications of AI in an increasingly algorithmic economy.
The narrative synthesis of peer-reviewed studies demonstrates that artificial intelligence has irrevocably reconfigured the foundations of strategic management. No longer a peripheral tool, AI now sits at the heart of organizational decision systems, simultaneously amplifying predictive capacity, reshaping resource orchestration, and redefining the very sources of competitive advantage [1, 2, 24, 28]. The five interlocking themes—AI-enabled decision systems, competitive advantage mechanisms, human–AI collaboration, algorithmic governance, and digital transformation—collectively portray a field in transition from conceptual optimism to evidence-based realism.
Key patterns emerge with striking clarity. Hybrid architectures outperform both purely human and purely algorithmic alternatives across operational, innovative, and strategic domains [2, 3, 6, 12, 22]. Competitive advantage is increasingly situated, data-dependent, and transient, demanding continuous organizational learning and governance vigilance rather than one-time technology adoption [24, 27-29]. Human judgment retains irreplaceable value in sense-making, ethical calibration, and creative integration, yet the rise of generative models introduces substitution pressures that demand proactive workforce and capability redesign [11, 13, 22]. Governance, once an afterthought, now constitutes a core strategic competence; organizations that embed ethical oversight and algorithmic transparency into their decision architectures secure not only legitimacy but also sustained differentiation [10, 23]. Finally, digital transformation emerges as the overarching process through which these elements coalesce, turning technological potential into measurable business value only when decision systems, collaboration models, and renewal capabilities are deliberately aligned [4, 5, 15, 17].
These insights carry profound implications for practice. Executives must move beyond pilot projects to architect enterprise-wide hybrid decision systems supported by board-level AI governance frameworks. Investment priorities should shift from raw computational power toward complementary organizational routines—training programs that foster human–AI symbiosis, transparency protocols that mitigate bias, and dynamic capability-building processes that convert algorithmic insights into strategic action. Managers in SMEs and emerging markets, in particular, require tailored pathways that account for resource constraints while still capturing AI-driven gains. Policymakers, meanwhile, face the challenge of crafting regulatory environments that promote innovation without stifling the very experimentation that generates competitive advantage.
For scholars, the review signals fertile ground for theoretical advancement. The resource-based view, dynamic capabilities, and behavioral strategy literatures all require updating to accommodate AI as both a resource and decision agent [20, 21, 28]. Micro-foundational research is urgently needed to unpack how individual managers and teams interact with algorithmic recommendations under varying levels of uncertainty and accountability. Interdisciplinary bridges—linking strategic management with information systems ethics, organizational learning, and innovation policy—promise richer explanations of the multi-level mechanisms at play. Longitudinal, multi-industry, and cross-cultural designs will be essential to test the durability of observed patterns and to capture the evolving impact of successive AI generations, particularly generative models whose strategic implications are only beginning to surface [11, 13].
The narrative also underscores important cautions. Over-reliance on opaque systems risks deskilling, ethical lapses, and strategic brittleness [10, 23]. Competitive advantages built solely on data scale or algorithmic speed prove fragile in fast-diffusing technological landscapes [24, 28]. Organizations that treat AI as a plug-and-play solution rather than a systemic transformation catalyst will likely experience short-term gains followed by long-term erosion of both performance and legitimacy. The field’s evolution from 2017 conceptual explorations of symbiosis [21, 22] to 2024 evidence of situated, contingent, and ethically fraught outcomes [11, 13, 24, 28] therefore serves as both roadmap and warning.
Ultimately, this review portrays strategic management in the AI era as an interconnected paradigm in which technologies, decision systems, human agency, and governance co-evolve. Competitive advantage accrues to those organizations that master this interplay—building adaptive, transparent, and ethically grounded decision architectures that amplify rather than supplant human strategic judgment. By synthesizing the most influential streams in the literature, mapping their interdependencies, and spotlighting persistent gaps, the manuscript provides a structured foundation for scholars and a practical compass for executives navigating the complexities of the digital age. The journey ahead demands continued rigorous inquiry, cross-disciplinary dialogue, and courageous organizational experimentation. Only through such collective effort can the promise of artificial intelligence translate into enduring strategic value for firms and society alike.
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
