Leading Algorithmic Authority: Why Ethical AI Governance Depends on Legitimacy Infrastructure, Not Compliance Checklists

8 hours ago

17 min read

Listen to a review of this article:

Abstract: Ethical AI governance has become a strategic imperative as algorithmic systems increasingly mediate consequential organizational decisions affecting credit access, service delivery, and social participation. Existing frameworks emphasize principles and technical assurance but assume stable infrastructure, coherent institutions, and baseline trust—conditions that rarely hold in volatile environments. This article reconceptualizes ethical AI governance as legitimacy infrastructure: a leadership-designed capability system enabling organizations to deploy algorithmic authority while sustaining contestability, accountability, and procedural justice when external conditions are unstable. Drawing on legitimacy theory and leadership scholarship, the article introduces a three-dimensional volatility typology—infrastructural, institutional, and socio-political—and proposes a Sensing–Stabilizing–Legitimizing (SSL) leadership framework. The analysis demonstrates that under volatility, ethical governance succeeds only when leaders institutionalize legitimacy production rather than rely on documentation alone. Organizations must build redundancy into harm detection, treat governance documentation as adaptive rather than static, and prioritize procedural justice mechanisms that make algorithmic decisions genuinely contestable. The framework offers actionable guidance for leaders navigating the dual pressures of innovation acceleration and disruption management in an era where algorithmic authority increasingly shapes organizational power.

Algorithmic systems have evolved from operational tools into decision infrastructures that fundamentally reshape how organizations exercise authority. Credit scoring determines financial access. Fraud detection systems classify risk at scale. Hiring algorithms filter opportunity. Welfare targeting systems allocate scarce public resources. Clinical decision support influences treatment pathways. These systems do not merely automate existing processes—they create new regimes of visibility, classification, and exclusion that demand active governance and clear leadership accountability.

The transformation is profound: algorithmic authority now mediates life chances through computational judgments that operate at speeds and scales beyond traditional human oversight. When these systems fail, the consequences extend far beyond technical errors—they manifest as systematic exclusion, unexplained denial, and compounded disadvantage for already marginalized populations. The governance challenge is therefore not peripheral to digital transformation but central to it: organizations must determine how to justify, constrain, and maintain accountability for this new form of power.

Contemporary leadership faces dual pressures that make this challenge particularly acute. Innovation acts as a pull factor—competitive advantage, efficiency gains, and service scalability drive rapid AI adoption across sectors. Simultaneously, disruption functions as a push factor—regulatory volatility, infrastructural fragility, and legitimacy crises expose organizations to governance failures that trigger reputational collapse and regulatory sanctions. Leaders must implement AI systems to remain competitive while managing institutional uncertainty, technological fragility, and contested social consent. Ethical AI governance thus becomes the capability determining whether innovation remains sustainable or becomes self-destructive.

Despite rapid progress in AI ethics scholarship, existing frameworks inadequately address this volatility-governance nexus. Responsible AI frameworks specify normative goals—fairness, accountability, transparency—but assume stable infrastructure and coherent regulatory institutions. AI assurance frameworks provide verification mechanisms through documentation, auditing, and monitoring, but presuppose consistent data pipelines, enforceable standards, and accessible recourse. Regulatory frameworks such as the EU AI Act demonstrate significant progress, yet implementation debates confirm that institutional environments remain unpredictable. Where stability conditions hold, ethical AI governance can be framed as policy alignment plus technical controls. Where they fail—as they frequently do across much of the global South and increasingly elsewhere—governance frameworks lose their operational footing.

This article argues that ethical AI governance under volatility is fundamentally a leadership problem of legitimacy production, not a technical problem of compliance. Where stability is limited, ethics is not a peripheral constraint on strategy; it becomes a necessary precondition for strategy to be executable. Leaders who treat ethical AI as a delegated technical matter often discover the ethics problem returning as strategic crisis: adoption collapses, trust erodes, and social actors challenge the organization's right to decide.

The Algorithmic Authority Landscape

Defining Algorithmic Authority in Contemporary Organizations

The term "ethical AI" implies a moral overlay applied to technical systems, as if ethics represents constraints imposed from outside. The governance challenge becomes analytically clearer when we shift focus to algorithmic authority: the capacity of automated systems to classify, rank, predict, recommend, and decide in ways that shape outcomes for individuals and communities. This authority is not merely technical—it is deeply institutional, embedded in organizational routines, vendor relationships, regulatory expectations, and public interpretations of fairness.

Algorithmic authority reorganizes organizational power by distributing consequences through computational procedures that are difficult to scrutinize and harder to contest. Approval or denial, inclusion or exclusion, trust or suspicion—these determinations increasingly flow from automated systems rather than human judgment alone. The shift represents a fundamental change in how organizations exercise power: decisions that once required human deliberation, documentation, and potential reconsideration now occur at machine speed, often without meaningful human oversight or accessible explanation.

Prevalence and Sectoral Distribution

Algorithmic decision-making has achieved widespread adoption across critical domains. In financial services, credit scoring algorithms determine access to capital, with models trained on historical data that may encode systematic biases against underserved populations. Healthcare systems deploy clinical decision support tools that influence diagnostic pathways and treatment recommendations, creating new questions about medical authority and accountability when algorithmic suggestions diverge from clinical judgment.

Public sector organizations have embraced algorithmic systems for welfare targeting, fraud detection, and resource allocation—domains where errors concentrate harm among vulnerable populations least equipped to contest decisions. Platform companies deploy content moderation algorithms that shape political discourse and cultural expression at global scale, making consequential determinations about what speech is visible, suppressed, or removed entirely. Human resources systems use algorithmic screening to filter job applicants, raising persistent concerns about proxy discrimination and the perpetuation of workplace inequality.

The common thread across these domains is consequentiality: algorithmic systems increasingly mediate access to fundamental goods—financial services, healthcare, social support, employment, and public participation. When these systems fail or reflect embedded biases, the harms are not merely technical errors but systematic injustices that undermine trust, perpetuate exclusion, and concentrate disadvantage.

Drivers of Accelerating Adoption

Several forces propel organizations toward algorithmic decision-making despite governance uncertainties. Competitive pressure drives adoption as early movers gain efficiency advantages that force industry-wide responses. Cost reduction through automation creates powerful internal constituencies favoring deployment even when governance infrastructure remains immature. Scale imperatives push organizations toward algorithmic systems when manual processes cannot keep pace with growth.

Data accumulation creates internal momentum: organizations that have invested in data infrastructure face pressure to extract value through analytics and prediction. Vendor ecosystems actively promote AI adoption, often emphasizing capability while understating governance complexity. Regulatory encouragement in some jurisdictions positions AI as advancing policy goals such as financial inclusion or healthcare access, creating institutional support for rapid deployment.

Perhaps most significantly, the perception that AI adoption is inevitable—that competitive survival requires embracing algorithmic systems—drives organizations to deploy systems before governance capabilities fully mature. This "innovation imperative" creates the conditions for governance failures when deployment velocity exceeds the organizational capacity to detect harm, stabilize safeguards, and sustain legitimacy.

Organizational and Stakeholder Consequences of Algorithmic Authority

Organizational Performance and Strategic Impacts

The consequences of algorithmic authority failures extend far beyond isolated incidents. Organizations experience multiple forms of strategic damage when governance proves inadequate. Reputational collapse can occur rapidly when algorithmic harms become public, particularly when systems appear to encode discrimination or exclude vulnerable populations. Civil society organizations, investigative journalists, and advocacy groups increasingly scrutinize algorithmic systems, with findings disseminated through social media at speeds that overwhelm traditional crisis management.

Regulatory intervention represents another significant risk. As governments strengthen oversight of AI systems, organizations face investigations, sanctions, and enforcement actions when algorithmic decisions produce systematic harms. The EU AI Act, though still evolving in implementation, signals a shift toward mandatory governance standards with meaningful penalties for non-compliance. Organizations that treated ethical AI as voluntary risk management face unexpected legal exposure as regulatory expectations crystallize.

Operational disruption occurs when algorithmic failures force systems offline or require manual intervention at scale. Organizations discover that automated decision-making creates dependencies: when algorithmic systems fail, the human capacity to process decisions manually often no longer exists. Service delivery halts, backlogs accumulate, and organizations cannot easily revert to pre-algorithmic processes.

Market exclusion can follow governance failures as clients, partners, or platforms withdraw from organizations perceived as unable to manage algorithmic authority responsibly. In sectors with concentrated buyer power—such as public procurement or enterprise software—governance failures can eliminate organizations from consideration regardless of technical capability.

Individual and Community Impacts

The human consequences of algorithmic authority failures are often severe and concentrated among already marginalized populations. Exclusion-by-algorithm occurs when automated systems systematically deny access to services, credit, or opportunities based on patterns embedded in training data. Individuals experience these denials as arbitrary and inexplicable—credit applications rejected without clear reasons, job applications filtered out by opaque criteria, social services denied through automated eligibility determinations.

Compounded disadvantage emerges when algorithmic exclusion reinforces existing inequalities. Those with limited digital access generate fewer data traces, making them simultaneously more likely to be excluded by algorithmic systems and less able to contest decisions. Communities already underserved by formal institutions face additional barriers when algorithmic systems encode historical patterns of exclusion as predictive features.

Procedural injustice—the experience of facing consequential decisions without explanation, recourse, or human accountability—erodes trust in institutions. When individuals cannot understand why they were denied, whom to appeal to, or how to correct errors, algorithmic authority feels arbitrary and dehumanizing. This erosion of procedural justice has broader social consequences: it undermines confidence in institutions, fuels opposition to technological change, and creates political openings for regulatory backlash.

Dignity harms occur when individuals are reduced to data profiles and subjected to classifications they find offensive or inaccurate. Being labeled high-risk by an algorithm, flagged as potentially fraudulent, or categorized in ways that conflict with self-understanding creates psychological harm distinct from material exclusion. These dignity harms are particularly acute when algorithmic classifications touch on sensitive attributes—health status, creditworthiness, or perceived behavioral patterns.

Evidence-Based Organizational Responses

Table 1: Algorithmic Authority Governance Practices and Examples

Sector or Organization	Governance Mechanism	Description of Practice	Target Outcome	Key Stakeholders Involved	Practical Examples
Amsterdam	Algorithmic Registry	A public registry documenting the purpose, impact assessments, and monitoring protocols of AI systems used in municipal services.	Accountability, transparency, and crisis prevention	Municipal service users, city officials, and public stakeholders	Public disclosure of AI systems used in municipal services.
Microsoft	Aether Committee and Impact Assessments	Embedded stakeholder engagement throughout the AI development lifecycle, utilizing a cross-disciplinary oversight committee and publishing impact assessments for high-stakes systems.	Legitimacy and sustained engagement	Cross-disciplinary experts, civil society organizations, and external stakeholders	Maintaining ongoing dialogue with civil society regarding AI ethics.
Unilever	Fairness Task Force	A distributed governance model that conducts quarterly reviews of system performance and evaluates bias metrics in hiring systems.	Equity standards and operational relevance of ethics	Data scientists, HR professionals, legal counsel, and external advisors	Authority to modify or suspend hiring tools that fail equity standards.
Financial Services	Human Override Protocols	Specifying conditions under which human judgment supersedes algorithmic recommendations in high-stakes decisions like credit scoring.	Procedural justice and contestability	Human decision-makers and affected applicants	Escalation pathways for credit applications rejected by algorithms.
DBS Bank (Singapore)	Customer-facing AI Transparency Framework	Providing clear explanations for algorithmic lending decisions and establishing human-led escalation pathways for disputes and appeals.	Trust, detection of errors, and corrective accessibility	Customers and human decision-makers	Algorithmic lending decision explanations and human oversight for appeals.
Capital One	AI Governance Office	Dedicated office for continuous environmental scanning of regulatory developments and translating them into operational guidance.	Governance resilience and regulatory compliance	Regulatory bodies, operational teams, and governance officers	Active management of regulatory volatility through adaptive documentation.

Transparent Communication and Stakeholder Engagement

Effective governance begins with recognizing that algorithmic systems require ongoing communication, not one-time explanations. Organizations that sustain legitimacy treat transparency as a continuous leadership obligation rather than a technical documentation requirement. This means establishing regular stakeholder engagement mechanisms—advisory boards, community consultations, and feedback channels—that bring external perspectives into governance decisions before deployment.

Practical approaches include:

Stakeholder mapping and engagement planning that identifies who is affected by algorithmic decisions and creates appropriate participation mechanisms
Plain-language explanations of algorithmic decision logic that avoid technical jargon while providing meaningful insight into how systems operate
Regular reporting on algorithmic outcomes disaggregated by demographic characteristics to surface disparate impacts before they become crises
Accessible complaint mechanisms that allow affected individuals to flag concerns without requiring technical expertise
Governance transparency through public disclosure of algorithmic impact assessments, audit findings, and remediation actions

Microsoft has embedded stakeholder engagement throughout its AI development lifecycle, establishing the Aether Committee to provide cross-disciplinary oversight and creating mechanisms for external input on AI ethics. The company publishes impact assessments for high-stakes AI systems and maintains ongoing dialogue with civil society organizations, recognizing that legitimacy requires sustained engagement rather than episodic consultation.

Procedural Justice and Contestability Mechanisms

Legitimacy depends fundamentally on procedural justice: the perception that decision processes are fair, regardless of outcomes. For algorithmic systems, procedural justice requires that affected individuals can understand decisions, challenge determinations, and access meaningful recourse. Organizations that build procedural justice into system design rather than adding it as an afterthought create more resilient governance infrastructure.

Effective procedural justice mechanisms include:

Decision explanations tailored to user literacy levels, providing both the outcome and meaningful insight into contributing factors
Clear escalation pathways that route challenges to appropriately empowered human decision-makers with authority to override algorithmic determinations
Timely response commitments that prevent appeals from languishing in bureaucratic queues
Remediation tracking that documents what happened to appeals and whether patterns suggest systemic issues
Human override protocols that specify conditions under which human judgment supersedes algorithmic recommendations

DBS Bank in Singapore created a customer-facing AI transparency framework that provides clear explanations for algorithmic lending decisions, establishes escalation pathways for disputes, and maintains human oversight for appeals. The bank recognized that trust depends not on algorithmic perfection but on customers' confidence that errors can be detected, explained, and corrected through accessible processes.

Algorithmic Impact Assessment and Monitoring

Pre-deployment and ongoing impact assessments represent essential governance infrastructure, yet many organizations treat them as compliance formalities rather than living governance instruments. Effective impact assessments explicitly evaluate who is harmed when systems fail, particularly focusing on vulnerable populations most likely to experience exclusion. Ongoing monitoring extends beyond technical performance metrics to track equity indicators, complaint patterns, and qualitative feedback that signal emerging legitimacy challenges.

Robust assessment and monitoring practices include:

Pre-deployment risk analysis that maps potential harms across stakeholder groups before systems go live
Equity metrics tracking disparate impact across demographic characteristics throughout the system lifecycle
Data quality assessments that examine whether training data represents the populations subject to algorithmic decisions
Drift detection that identifies when system performance changes over time or across contexts
Qualitative feedback integration that supplements quantitative metrics with user experiences and community concerns

The city of Amsterdam developed an algorithmic registry that publicly documents AI systems used in municipal services, including their purpose, impact assessments, and monitoring protocols. The registry treats transparency as governance infrastructure: by making algorithmic systems visible, the city creates accountability and enables stakeholders to identify concerns before they escalate into crises.

Organizational Capability Building

Sustainable governance requires distributed capability, not centralized control. Organizations that rely on small ethics teams or external consultants to manage algorithmic authority discover that governance cannot scale when responsibility is isolated rather than embedded. Effective capability building creates shared competencies across technical, operational, and leadership roles so that ethical considerations shape design choices, deployment decisions, and operational responses.

Capability-building approaches include:

Cross-functional governance councils with authority to delay or modify deployment when ethical boundaries are crossed
Embedded ethics roles within product and engineering teams rather than isolated compliance functions
Leadership training that positions algorithmic governance as strategic capability rather than technical specialization
Escalation protocols that empower operational staff to pause systems when harms emerge
Decision rights frameworks that clarify who can authorize deployment, who owns risk acceptance, and who must be consulted before high-stakes systems go live

At Unilever, the company established a "Fairness Task Force" comprising data scientists, HR professionals, legal counsel, and external advisors to oversee algorithmic hiring systems. The task force conducts quarterly reviews of system performance, evaluates bias metrics, and has authority to modify or suspend tools that fail equity standards. This distributed governance model ensures ethical considerations remain operationally relevant rather than relegated to policy documents.

Adaptive Documentation and Regulatory Sensing

In volatile institutional environments, governance documentation cannot be static. Regulations evolve, enforcement priorities shift, and compliance expectations change faster than traditional policy update cycles. Organizations that treat documentation as adaptive instruments—regularly updated to reflect regulatory shifts, stakeholder concerns, and operational realities—build governance resilience that withstands institutional volatility.

Adaptive documentation practices include:

Living impact assessments that are updated when systems change, contexts shift, or new harms emerge
Regulatory monitoring functions that track policy developments across jurisdictions and translate implications for operational teams
Stakeholder feedback integration that treats external concerns as inputs requiring documented responses
Version control and change tracking that creates audit trails showing how governance documentation evolves
Scenario planning that anticipates regulatory directions and prepares governance responses before requirements crystallize

Capital One established a dedicated AI governance office that maintains continuous environmental scanning for regulatory developments, translates emerging requirements into operational guidance, and ensures governance documentation reflects current expectations rather than outdated standards. The office treats regulatory volatility as a permanent condition requiring active management rather than exceptional disruption.

Building Long-Term Governance Capacity

Leadership as Institutional Sensemaker

Effective governance under volatility requires leaders who function as institutional sensemakers—scanning, interpreting, and translating signals from regulatory, social, and technical environments into actionable governance responses. This sensemaking work is strategic, not administrative. It shapes risk appetite, influences investment decisions, and determines whether organizations treat emerging governance challenges as opportunities for capability building or threats to be minimized.

Leaders who excel at institutional sensemaking establish routines that surface weak signals before they become crises. They create channels for frontline staff to escalate concerns without fear of punishment. They engage with civil society organizations, regulators, and academic researchers to access perspectives that internal dashboards cannot provide. They treat governance ambiguity as information to be processed rather than noise to be filtered.

This sensemaking capability proves particularly valuable in volatile institutional environments where regulatory expectations shift rapidly and enforcement priorities change without warning. Organizations led by effective sensemakers adapt proactively rather than reactively, updating governance practices before external pressure forces change.

Architecting Adaptive Guardrails

Governance structures must balance two competing imperatives: providing clear constraints that prevent harm while remaining flexible enough to accommodate learning and adaptation. Organizations that establish overly rigid governance frameworks discover that rules become obstacles—teams work around controls or governance becomes performative rather than operational. Organizations with excessively flexible governance risk drift—principles remain aspirational but fail to constrain decisions when trade-offs arise.

Adaptive guardrails resolve this tension by specifying governance outcomes—decisions must be contestable, systems must be monitored, harms must trigger responses—while allowing operational discretion in how outcomes are achieved. This approach embeds governance into decision processes rather than treating it as external review. Teams understand what governance requires but maintain agency in implementation.

Effective adaptive guardrails include clear decision rights, escalation triggers that activate when systems approach ethical boundaries, and authorization protocols that specify conditions under which deployment requires senior leadership approval. These structures create governance that operates in real time rather than after deployment when corrective action proves more costly and disruptive.

Sustaining Legitimacy Through Procedural Justice

Long-term legitimacy depends on more than technical correctness—it requires that affected stakeholders perceive algorithmic authority as appropriately constrained, transparently exercised, and genuinely accountable. Organizations sustain this perception by institutionalizing procedural justice: ensuring that decisions are explainable, mistakes are acknowledged and corrected, and power remains contestable.

Procedural justice operates through multiple mechanisms. Transparent decision logic allows affected individuals to understand why outcomes occurred. Accessible appeals create pathways for challenging determinations. Timely responses demonstrate that challenges receive genuine consideration rather than bureaucratic dismissal. Visible accountability—public disclosure of governance actions, audit findings, and remediation steps—signals that organizations take responsibility seriously.

Organizations that invest in procedural justice infrastructure discover that legitimacy proves more durable than organizations relying primarily on technical optimization. When systems fail—as all complex systems eventually do—procedural justice provides resilience: stakeholders remain willing to engage because they believe errors will be acknowledged and corrected rather than concealed or dismissed.

Conclusion

Algorithmic authority represents a fundamental shift in how organizations exercise power, creating new regimes of classification, exclusion, and opportunity allocation that demand active governance and clear leadership accountability. Ethical AI governance cannot be treated as a technical compliance matter delegated to specialized teams. Under conditions of volatility—where infrastructure is unstable, institutions are shifting, and social consent is contested—governance becomes a core strategic leadership capability determining whether innovation remains sustainable or becomes self-destructive.

This article has argued that effective governance under volatility requires reconceptualizing ethical AI as legitimacy infrastructure: a leadership-designed capability system that sustains contestability, accountability, and procedural justice when external conditions are unstable. The Sensing–Stabilizing–Legitimizing framework specifies the leadership work required to make existing governance instruments—documentation, monitoring, impact assessments—durable under volatility rather than merely aspirational.

The practical implications are clear. Leaders must institutionalize harm detection through redundant sensing mechanisms that surface exclusion patterns before they become crises. Organizations must treat governance documentation as adaptive instruments that evolve with regulatory shifts rather than static compliance artifacts. Procedural justice mechanisms—accessible explanations, escalation pathways, and meaningful recourse—must be embedded into system design rather than added as afterthoughts.

The deeper insight is that legitimacy is not a byproduct of good governance but its precondition. Organizations cannot govern what they have not first legitimized. As algorithmic authority continues to expand in scope and consequence, the capacity to construct and sustain legitimacy infrastructure becomes the defining leadership capability of the algorithmic era—determining not only organizational success but whether technological power remains socially acceptable and democratically accountable.

Research Infographic

Architecting Algorithmic Legitimacy Slide Deck

References

Bamberger, P. A., & Pratt, M. G. (2010). Moving forward by looking back: Reclaiming unconventional research contexts and samples in organizational scholarship. Academy of Management Journal, 53(4), 665–671.
Bennett, N., & Lemoine, G. J. (2014). What a difference a word makes: Understanding threats to performance in a VUCA world. Business Horizons, 57(3), 311–317.
Bitektine, A. (2011). Toward a theory of social judgments of organizations: The case of legitimacy, reputation, and status. Academy of Management Review, 36(1), 151–179.
Crawford, K. (2021). Atlas of AI: Power, politics, and the planetary costs of artificial intelligence. Yale University Press.
Deephouse, D. L., Bundy, J., Tost, L. P., & Suchman, M. C. (2017). Organizational legitimacy: Six key questions. In R. Greenwood, C. Oliver, T. B. Lawrence, & R. E. Meyer (Eds.), The SAGE handbook of organizational institutionalism (2nd ed., pp. 27–54). SAGE Publications.
Edelman, L. B. (2016). Working law: Courts, corporations, and symbolic civil rights. University of Chicago Press.
Eubanks, V. (2018). Automating inequality: How high-tech tools profile, police, and punish the poor. St. Martin's Press.
European Commission. (2024). Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence. Official Journal of the European Union.
Floridi, L., Cowls, J., Beltrametti, P., & Vayena, E. (2018). AI4People—An ethical framework for a good AI society: Opportunities, risks, principles, and recommendations. Minds and Machines, 28(4), 689–707.
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Daumé, H., & Crawford, K. (2021). Datasheets for datasets. Communications of the ACM, 64(12), 86–92.
George, G., Corbishley, C., Khayesi, J. N. O., Haas, M. R., & Tihanyi, L. (2016). Bringing Africa in: Promising directions for management research. Academy of Management Journal, 59(2), 377–393.
Graham, S. (2011). Disrupted cities: When infrastructure fails. In S. Graham (Ed.), Disrupted cities: When infrastructure fails (pp. 1–2). Routledge.
Hällgren, M., Rouleau, L., & de Rond, M. (2018). A matter of life or death: How extreme context research matters for management and organization studies. Academy of Management Annals, 12(1), 111–153.
Heifetz, R., Grashow, A., & Linsky, M. (2009). The practice of adaptive leadership: Tools and tactics for changing your organization and the world. Harvard Business Press.
Jobin, A., Ienca, M., & Vayena, E. (2019). The global landscape of AI ethics guidelines. Nature Machine Intelligence, 1(9), 389–399.
Kroll, J. A., Huey, J., Barocas, S., Felten, E. W., Reidenberg, J. R., Robinson, D. G., & Yu, H. (2017). Accountable algorithms. University of Pennsylvania Law Review, 165(3), 633–705.
Lawrence, T. B., Suddaby, R., & Leca, B. (Eds.). (2011). Institutional work: Actors and agency in institutional studies of organizations. Cambridge University Press.
MacLean, T. L., & Behnam, M. (2010). The dangers of decoupling: The relationship between compliance programs and unethical behavior. Academy of Management Journal, 53(6), 1499–1520.
Mair, J., Martí, I., & Ventresca, M. J. (2012). Building inclusive markets in rural Bangladesh: How intermediaries work institutional voids. Academy of Management Journal, 55(4), 819–850.
Maitlis, S., & Christianson, M. (2014). Sensemaking in organizations: Taking stock and moving forward. Academy of Management Annals, 8(1), 57–125.
Martin, K., & Waldman, A. (2023). Are algorithmic decisions legitimate? The effect of process and outcomes on perceptions of legitimacy of AI decisions. Journal of Business Ethics, 183(3), 653–670.
Mitchell, M., Wu, S., Zaldivar, A., Barnes, P., Vasserman, L., Hutchinson, B., … Gebru, T. (2019). Model cards for model reporting. Proceedings of the Conference on Fairness, Accountability, and Transparency (pp. 220–229). Association for Computing Machinery.
Mittelstadt, B. (2019). Principles alone cannot guarantee ethical AI. Nature Machine Intelligence, 1(11), 501–507.
National Institute of Standards and Technology. (2023). Artificial intelligence risk management framework (AI RMF 1.0). U.S. Department of Commerce.
Noble, S. U. (2018). Algorithms of oppression: How search engines reinforce racism. New York University Press.
Pasquale, F. (2015). The black box society: The secret algorithms that control money and information. Harvard University Press.
Raji, I. D., Smart, A., White, R. N., Mitchell, M., Gebru, T., Hutchinson, B., & Barnes, P. (2020). Closing the AI accountability gap: Defining an end-to-end framework for internal algorithmic auditing. Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency (pp. 33–44). Association for Computing Machinery.
Schoemaker, P. J. H., Heaton, S., & Teece, D. (2018). Innovation, dynamic capabilities, and leadership. California Management Review, 61(1), 15–42.
Scott, W. R. (2014). Institutions and organizations: Ideas, interests, and identities (4th ed.). SAGE Publications.
Selbst, A. D., Boyd, D., Friedler, S. A., Venkatasubramanian, S., & Vertesi, J. (2019). Fairness and abstraction in sociotechnical systems. Proceedings of the Conference on Fairness, Accountability, and Transparency (pp. 59–68). Association for Computing Machinery.
Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571–610.
Taeihagh, A. (2021). Governance of artificial intelligence. Policy and Society, 40(2), 137–157.
Teece, D. J., Peteraf, M., & Leih, S. (2016). Dynamic capabilities and organizational agility: Risk, uncertainty, and strategy in the innovation economy. California Management Review, 58(4), 13–35.
Uhl-Bien, M., & Arena, M. (2018). Leadership for organizational adaptability: A theoretical synthesis and integrative framework. The Leadership Quarterly, 29(1), 89–104.
Weick, K. E. (1995). Sensemaking in organizations. SAGE Publications.
Zietsma, C., & Lawrence, T. B. (2010). Institutional work in the transformation of an organizational field: The interplay of boundary work and practice work. Administrative Science Quarterly, 55(2), 189–221.

Jonathan H. Westover, PhD, Chief Research Officer (Nexus Institute for Work and AI); Co-Founder & Chief Workforce and Learning Officer (Future State University); Founder & CEO (Human Capital Innovations); Professor of Organizational Leadership & Change (UVU). Read Jonathan Westover's executive profile here.

Suggested Citation: Westover, J. H. (2026). Leading Algorithmic Authority: Why Ethical AI Governance Depends on Legitimacy Infrastructure, Not Compliance Checklists. Human Capital Leadership Review, 39(1). doi.org/10.70175/hclreview.2020.39.1.7