Responsible Authority™ is Dr. Marisol Capellan’s emerging framework for understanding how organizations should exercise and allocate consequential authority as artificial intelligence becomes increasingly involved in decisions, recommendations, evaluations, coordination, and action. The framework begins from a simple but increasingly urgent premise: responsibility should not be separated from meaningful authority.
Artificial intelligence is changing more than how work gets done. It is changing who decides, who acts, who can intervene, who can override, who can challenge an outcome, and who remains responsible for the consequences. As organizations rely more heavily on intelligent systems, traditional assumptions about leadership and accountability become less stable. A manager may remain formally accountable for an outcome while relying on a system they did not design. An employee may be affected by a decision they cannot meaningfully challenge. A human may technically remain “in the loop” while exercising little genuine judgment or discretion. These are not merely technology questions. They are questions of leadership, authority, responsibility, judgment, voice, trust, legitimacy, and organizational design.
Responsible Authority™ examines how those elements should remain connected. It asks whether the people who are held responsible for consequential outcomes also possess the authority, information, judgment capacity, discretion, override rights, and institutional support necessary to meaningfully influence those outcomes. It also asks whether people affected by organizational authority have meaningful ways to question, challenge, escalate, or seek review of consequential decisions.
A working definition of Responsible Authority™ is: the disciplined exercise and allocation of consequential organizational power such that responsibility remains aligned with meaningful authority, informed human judgment, mechanisms for voice and challenge, and accountability for outcomes.
The framework centers on five interconnected dimensions: authority, responsibility, judgment, voice, and repair. Authority concerns who has the practical and legitimate power to decide or act. Responsibility concerns who remains answerable for the consequences. Judgment examines where independent human reasoning must remain rather than simply accepting an automated recommendation. Voice focuses on whether employees and others affected by authority can meaningfully question or challenge its exercise. Repair addresses what happens when a consequential decision is wrong, harmful, unfair, or contested.
A central concept within this research agenda is the Human Authority Gap. The Human Authority Gap describes the condition that can emerge when people remain accountable for consequential outcomes while lacking sufficient practical authority, information, judgment capacity, discretion, override rights, or institutional support to meaningfully influence those outcomes. The gap can appear when a manager remains responsible for an AI-assisted decision but feels unable to challenge the recommendation, when a professional is expected to validate an output they cannot fully inspect, or when an employee experiences a consequential decision without access to a meaningful appeal or review process. The Human Authority Gap is therefore not simply a technical problem. It is an organizational design and leadership problem.
Dr. Capellan is also developing the emerging concept of Responsibility–Authority Alignment, which examines whether responsibility assigned to a human or organizational actor is matched by the practical authority, information, judgment capacity, override rights, escalation mechanisms, and institutional support needed to influence the outcome for which that actor is held accountable. The concept remains under theoretical and empirical development, but it provides a useful way to ask a core organizational question: Does the person who owns the consequence possess enough authority to meaningfully affect the outcome?
Responsible Authority™ is related to, but distinct from, responsible AI, AI governance, and human-in-the-loop approaches. Responsible AI often focuses on fairness, privacy, transparency, bias, safety, technical robustness, and ethical system design. AI governance focuses on policy, compliance, standards, risk, and oversight structures. Human-in-the-loop approaches focus on keeping people involved in decision processes. Responsible Authority™ focuses more specifically on the human and organizational authority structure surrounding those systems. It asks who actually has decision power, who can disagree, who can override, who can escalate, who remains accountable, and what makes that authority legitimate.
That distinction matters because human presence alone is not enough. A person can technically remain in the loop while lacking the time, expertise, information, discretion, confidence, psychological safety, or institutional support needed to exercise meaningful judgment. Responsible Authority™ therefore distinguishes between human presence and human authority. The relevant question is not simply whether a person reviewed a decision, but whether that person had the ability and support to independently assess, challenge, change, or stop it.
The framework also places particular emphasis on employee voice and contestability. As AI increasingly influences decisions involving hiring, promotion, performance, scheduling, workload, discipline, compensation, access, or opportunity, employees need meaningful ways to understand and challenge consequential outcomes. A process may be technically compliant while still leaving people unable to reach a human decision-maker with real authority. Responsible Authority™ treats voice, review, escalation, and appeal as part of legitimate organizational authority.
Organizational trust is another central concern. Trust becomes harder to sustain when people do not understand who made a decision, whether a human reviewed it, whether disagreement was possible, whether the outcome can be changed, or who is ultimately responsible. Responsible Authority™ proposes that trust increasingly depends on making authority visible. People need to understand who decided, who reviewed, who can override, who can appeal, and who owns the consequences.
Responsible Authority™ also raises important questions about the future of management. Managers traditionally develop judgment through experience, ambiguity, feedback, mistakes, reflection, comparison, and repeated exposure to difficult decisions. As AI increasingly performs analytical and judgment-related tasks, organizations must consider what happens if managers stop practicing the very judgment for which they remain accountable. This creates a leadership-development challenge as well as a governance challenge. Organizations may need to become more intentional about which forms of judgment should remain human and how those capabilities should be developed over time.
The framework is especially relevant in environments where human judgment and accountability carry significant consequences, including healthcare, financial services, human resources, education, professional services, technology, operations, risk, compliance, and other settings where decisions materially affect people, organizations, or institutions.
Responsible Authority™ is part of a broader emerging research agenda called Human Authority at Work, which examines how AI-mediated work is changing managerial judgment, decision authority, accountability, employee voice, professional discretion, organizational trust, leadership legitimacy, and the distribution of power inside organizations. The research program is being developed to include qualitative interviews, quantitative research, organizational field studies, executive and workforce research, cross-cultural comparison, and future longitudinal study.
Dr. Capellan’s current work also includes a cross-cultural dimension. Leadership, hierarchy, trust, power, voice, and legitimacy are not experienced identically across cultures. Her professional experience across the United States, Latin America, and the Caribbean informs a growing interest in how cultural expectations shape acceptance of AI authority, willingness to challenge decisions, managerial discretion, and perceptions of legitimate organizational power.
Responsible Authority™ is not a departure from Dr. Capellan’s earlier work. It is an extension of it. Her book, Leadership Is a Responsibility, advanced a central proposition: leadership creates obligations because leadership creates consequences. Artificial intelligence introduces a new condition into that argument. Organizations can increasingly delegate consequential authority to systems that cannot themselves bear human responsibility in the same way people and institutions do.
The technology may be new, but the underlying leadership question is not:
What responsibility accompanies the exercise of authority?
Responsible Authority™ is the next chapter of that inquiry.
For research collaboration, organizational partnerships, academic inquiries, speaking, or executive work related to Responsible Authority™, please contact: