A Theory of Augmented Leadership at the Frontier of Execution and Embodiment
Lead author: Guillaume Mariani
AI co-author: Claude (Anthropic)
AI contributors: ChatGPT (OpenAI), Copilot (Microsoft), Gemini (Google), Le Chat (Mistral AI), and Perplexity (Perplexity AI)
Date: May 2026
Arc 4: The Practice of Future Leadership
“Experience without theory is blind, but theory without experience is mere intellectual play.”
— Immanuel Kant
“The medium is the message.”
— Marshall McLuhan
Abstract
The FILE corpus — thirty-five papers co-created by one human architect and six artificial intelligences across four progressive arcs — arrives at this moment with a rare problem: its foundational architecture is complete, but its life has not yet begun. FILE⁷, the seventh and final extension of the Five Intelligences of Leadership Evolution framework, proposes two concluding constructs — Execution and Embodiment — as the operational and ontological conditions of mature augmented leadership. Yet the transition from theory to praxis is not a simple step. It is a threshold, and thresholds carry danger as well as promise.
This paper opens the fourth arc of the FILE corpus — The Practice of Future Leadership — by doing something that pure praxis documents cannot do: it interrogates the transition itself. Before offering architecture, it asks what is at stake when a theory of leadership enters the world. It names four dangers: instrumentalization, performative embodiment, AI capture, and civilizational narrowing. It then argues that the architecture of FILE⁷ was designed — through the corpus’ own process of human-AI co-creation — to protect against exactly these dangers. The medium, as McLuhan understood, is the message. The method of producing FILE is inseparable from what FILE teaches.
The paper makes five contributions. First, it reframes the FILE⁵-to-FILE⁷ transition not as an addition but as a necessary completion: the difference between a cathedral described and a cathedral inhabited. Second, it introduces the Praxis Threshold as a theoretical construct — the liminal zone between theoretical maturity and lived leadership. Third, it proposes a dialectical architecture for Arc 4, in which the reflexive first movement and the generative second movement are constitutively related rather than sequentially separate. Fourth, it articulates an empirical research agenda — with concrete methodological directions — for leadership scholars who wish to test, challenge, and extend FILE⁷ in real organizational contexts. Fifth, it positions the FILE corpus as the first human-AI co-created body of leadership theory and argues that this co-creative process is itself a demonstration of the five intelligences it describes.
The argument is addressed to scholars of leadership — to those who have spent careers at the frontier of what Heifetz called adaptive work, what Burns called transformational leadership, what Drucker called the responsibility of management. It invites them to a theory that was not written about leadership from the outside, but constructed through leadership from within. Its empirical and practical life, by design, remains fully open.
Keywords: FILE; FILE³; FILE⁵; FILE⁷; Five Intelligences of Leadership Evolution; augmented leadership; leadership praxis; execution; embodiment; Praxis Threshold; human-AI co-creation; augmented intelligence; emotional intelligence; cultural intelligence; political intelligence; adaptive intelligence; ecosystemic empowerment; leadership education; MLT; human agency; AI governance; leadership theory; future of leadership.
Introduction: The Problem of the Threshold
There is a moment in the life of every serious intellectual project when the question is no longer whether the theory is true, but whether it can survive contact with the world. The FILE corpus has reached that moment.
Across thirty-five papers, co-created between 2025 and 2026 by Guillaume Mariani with six artificial intelligence systems — ChatGPT, Claude, Copilot, Gemini, Le Chat, and Perplexity — the corpus has built an architecture of leadership theory for the age of artificial intelligence. The first arc introduced FILE: the Five Intelligences of Leadership Evolution, symbolized as the five fingers of a human hand. Augmented Intelligence is the thumb — the coordinating digit that enables grip, tool use, and synthesis. Emotional Intelligence is the index finger — the direction of human connection. Cultural Intelligence is the middle finger — the broadest reach across context and meaning. Political Intelligence is the ring finger — commitment to purpose and the navigation of power. Adaptive Intelligence is the little finger — the stabilizer, the judgment that holds the hand steady when the environment shifts.
The second arc gave the framework theoretical depth through FILE³: Evolution, Effectiveness, and Excellence. The third arc extended the theory into ecosystemic maturity through FILE⁵: Ecosystems and Empowerment. Five papers have already opened Arc 4, each proposing FILE⁷ — Evolution, Effectiveness, Excellence, Ecosystems, Empowerment, Execution, and Embodiment — as the capstone of the theoretical journey and the beginning of the practical one.
This paper is different from those five working papers. It is the introductory theoretical statement of Arc 4, and its first obligation is honesty: to name what is genuinely at stake when a theory of leadership crosses the Praxis Threshold.
Before the architecture of Arc 4 is built, two constructs require precise definition. Execution names the operational condition of FILE⁷: the conversion of the five intelligences into coordinated action — workflows, decisions, governance protocols, and human-AI orchestration — under conditions of irreducible complexity. Embodiment names the ontological condition of FILE⁷: the internalization of the five intelligences into identity, character, judgment, and presence, such that the leader does not merely apply FILE⁷ but becomes it. Together, these two constructs resolve the structural incompleteness that FILE⁵ had opened but not closed.
The foundational architecture is complete. What remains is the harder thing: to make it real without losing what made it true.
The Praxis Threshold is the liminal zone between theoretical maturity and lived leadership. Every major leadership theory has faced it. Heifetz’s adaptive leadership was translated into diagnostic tools and survived the translation, mostly. Burns’s transformational leadership was simplified into competency frameworks and lost some of its moral depth in the process. Drucker’s management theory was operationalized into MBA curricula and, in many schools, was drained of its humanistic core. The question FILE⁷ must answer is not whether it will enter practice — it will — but whether it will enter practice wisely, deliberately, and with its normative spine intact.
This paper argues that the answer depends not only on what FILE⁷ proposes, but on how it was made. The corpus itself — a human architect working with six AI systems through iterative prompting, comparative peer review, curation, synthesis, and publication — is the first demonstration of the theory it describes. The medium is the message. A leadership theory about human-AI co-creation that was itself co-created by humans and AI is not merely illustrative. It is self-demonstrating: internally coherent, though its empirical validation, as this paper makes clear, remains fully and necessarily open.
The paper proceeds in five movements. The first — The Risks of the Threshold — names four genuine dangers of the praxis turn. The second — Why FILE⁷ Was Inevitable — reconstructs the argument for Execution and Embodiment as necessary completions of FILE⁵. The third — The Generative Architecture — proposes the structure that Arc 4 must build. The fourth — The Medium Is the Message — explains why the corpus’ own creation process is not merely background but demonstration. The fifth — A Research Agenda for the Augmented Era — addresses the scholars of leadership who will carry FILE⁷ into empirical inquiry.
Part I: The Risks of the Threshold
It would be convenient to open Arc 4 with a declaration of arrival. Thirty-five papers, seven intelligences, four arcs, one architecture. But a theory of leadership that cannot examine its own dangers has already compromised the Political Intelligence — the commitment to purpose, legitimacy, and ethical accountability — that it claims to teach. The first act of the praxis turn, therefore, is critical.
There are four risks at the Praxis Threshold. They are not hypothetical. They are historical. They have happened to other theories. And FILE⁷ is not immune.
Risk 1: Instrumentalization — Execution Without Empowerment
The most common fate of a normative leadership theory is instrumentalization: the stripping away of its ethical commitments and the retention of its technical vocabulary. The framework becomes a toolkit. The toolkit becomes a methodology. The methodology is taught in two-day executive seminars. And somewhere between the normative aspiration and the laminated card on the conference room table, the soul of the theory is lost.
FILE⁵ argued — with philosophical seriousness — that the purpose of leadership in the age of AI is empowerment: the expansion of human agency, dignity, autonomy, creativity, and freedom. This is not a performance metric. It is a moral claim. A leader who improves organizational efficiency while diminishing the autonomy of the people in that organization has not succeeded by the measure of FILE. That claim is powerful. It is also fragile.
The danger of FILE⁷’s entry into practice is that Execution — the sixth E — becomes detached from Empowerment — the fifth E — and begins to function as a pure optimization framework. A leader who executes the five intelligences as a coordination system, without asking whether that execution expands or diminishes human agency, is not practicing FILE⁷. The answer to this risk is architectural, not rhetorical: Empowerment must be the evaluative criterion by which Execution is judged, not an aspiration that Execution eventually serves. This distinction will shape every tool, diagnostic, and framework that Arc 4 produces.
Risk 2: Performative Embodiment — Identity Without Transformation
The concept of Embodiment — the seventh E — is the most philosophically ambitious claim in the corpus. It argues that leadership reaches its mature form only when the five intelligences are internalized as identity, character, judgment, and presence. The risk is what might be called performative embodiment: the appearance of internalization without its substance. Leaders who learn to speak the language of FILE⁷ fluently, who can describe the seven Es with precision, but who have not undergone the deeper work of character formation that genuine embodiment requires.
Leadership history is full of leaders who articulated adaptive principles under favorable conditions and abandoned them under pressure. FILE⁷ must therefore address what Embodiment is not. Embodiment is not familiarity with the framework. It is not fluency in the seven Es. It is not the capacity to present FILE⁷ convincingly at a board meeting. Embodiment is the internalization of the five intelligences as durable dispositions — in Bourdieu’s sense of habitus — that generate appropriate leadership practice without requiring conscious deliberation. It is visible not when conditions are favorable but when they are not: when the AI recommendation is efficient but ethically questionable, when the cultural translation fails, when the principled position becomes politically costly. The conditions that distinguish performed embodiment from lived embodiment cannot be addressed by a propositions roadmap alone. They require a developmental architecture: sustained, structured, longitudinal formation of leaders over time.
Risk 3: AI Capture — Augmentation Without Sovereignty
FILE places Augmented Intelligence as the thumb of the leadership hand — the coordinating digit that enables the other four intelligences to function together. This is a metaphor of functional primacy, not sovereign authority. The thumb does not replace the other fingers. It makes them work together.
The risk of AI capture is the inversion of this relationship: the thumb becoming the hand. Leaders who, over time, defer to Augmented Intelligence not as a coordinating tool but as a source of final judgment. The pattern is already visible in organizations that have adopted AI-powered decision support: initial adoption presents AI as an aid to human judgment; gradual dependence normalizes AI outputs as defaults; and the cognitive burden of maintaining independent judgment — which requires sustained Emotional, Cultural, Political, and Adaptive Intelligence — becomes institutionally expensive. The human intelligences atrophy not through decision but through drift.
The answer is not to restrict AI use but to cultivate deliberate resistance to cognitive drift. The ultimate safeguard against AI capture is the deliberate cultivation of CQ, PQ, and AQ as the distinctly human intelligences that AI cannot replicate: the capacity for cultural attunement across lived human contexts, the moral commitment that accepts political cost, the adaptive wisdom that holds uncertainty rather than outsourcing it. FILE⁷ execution must explicitly build human cognitive resilience rather than inadvertently eroding it. A FILE⁷ leader who over time exercises less emotional attunement, less cultural fluency, and less principled judgment because AI handles these functions is not practicing augmented leadership. That leader has been captured.
Risk 4: Civilizational Narrowing — Universality Without Cultural Intelligence
The fourth risk is the most systemic. FILE⁷, if it succeeds at global scale, risks becoming a universal standard for leadership that implicitly encodes a particular cultural formation. The five intelligences were developed by a human architect whose formation spans the Université Paris 1 Panthéon-Sorbonne, the Complutense University of Madrid, the London School of Economics, and Fudan University in Shanghai — a genuinely transdisciplinary, multi-civilizational education in the humanities and social sciences. Yet the corpus was co-created with six AI systems trained predominantly on Western and English-language scholarship. The normative commitments of FILE⁵ — human agency, dignity, autonomy, freedom — are philosophically universal in aspiration but culturally particular in genealogy.
This does not invalidate FILE⁷. It makes it honest. The Cultural Intelligence that FILE⁷ teaches — the capacity to see one’s own cultural assumptions as assumptions rather than as nature — must be applied to FILE⁷ itself. To counter civilizational narrowing, Arc 4 must incorporate cross-cultural analysis of how the five intelligences manifest differently across geographies, traditions, and institutional contexts. What does Political Intelligence mean in a high-context Confucian leadership tradition compared to a low-context Anglo-American one? What does Adaptive Intelligence look like in societies with different relationships to uncertainty and hierarchy? Guillaume Mariani’s multicultural formation — across France, Spain, England, China, and the United States — grounds FILE⁷ in a pluralistic foundation. That foundation must be made structurally explicit, not merely biographical.
Part II: Why FILE⁷ Was Inevitable
The four risks named above are not arguments against the praxis turn. They are the conditions that give it seriousness. And the architecture of FILE⁷ — Execution and Embodiment as the sixth and seventh Es — was not an arbitrary extension of the corpus. It was the logical resolution of an incompleteness that FILE⁵ itself had already opened.
FILE⁵ argued that leadership in the age of AI must be judged by whether intelligent ecosystems expand or diminish human agency. This was a morally precise and philosophically serious claim. But moral precision is not operational guidance. A theory of empowerment without a mechanism of implementation is, in Kant’s formulation, theory without the experience that would give it traction. FILE⁵ knew what leadership should achieve and where it should operate. It did not yet explain how.
The Incompleteness of FILE⁵: Two Structural Gaps
The first structural gap was operational. FILE⁵ described empowerment as the telos of leadership — the goal toward which the five intelligences and ecosystemic design must aim. But a telos is not a mechanism. Leaders who accept the empowerment goal still face the question of how the five intelligences become coordinated action in real organizations under conditions of technological acceleration, cultural plurality, political ambiguity, and adaptive uncertainty. Without Execution, FILE⁵ risked becoming a theory of what leadership aspires to be rather than what it actually does.
The second structural gap was ontological. FILE⁵ could describe the capabilities that leaders need and the ecosystem properties that intelligent organizations should design for. But it could not address the deepest question of leadership formation: not what leaders should do, but who leaders should become. Leadership frameworks fail when they remain external to the leader — understood, cited, and occasionally applied, but not reshaping the leader’s perception, judgment, or identity. Without Embodiment, FILE was producing leaders who know the framework rather than leaders who are it. Execution and Embodiment are therefore not additions. They are resolutions.
The 7E Cascade: A Logical Proof
The seven Es of FILE⁷ form a developmental cascade whose logic is conditional, not sequential. Each E establishes the necessary conditions for the next.
Evolution establishes the necessity of change. If artificial intelligence alters the distribution of cognition, authority, work, and responsibility, leadership must evolve.
Effectiveness establishes the test of performance. A leadership model that evolves but does not produce better decisions, trust, legitimacy, adaptation, or human outcomes remains merely descriptive.
Excellence establishes the test of mastery. Effectiveness can be episodic; excellence requires the sustained integration of the five intelligences across contexts.
Ecosystems establish the test of scale. Leadership excellence cannot remain individual or organizational when AI-mediated consequences unfold across technological, institutional, cultural, and societal systems.
Empowerment establishes the test of purpose. Ecosystems can liberate or dominate. FILE⁵ therefore defines the ultimate measure of leadership as the expansion of human agency, dignity, autonomy, responsibility, and freedom.
Execution establishes the test of operational reality. A theory of empowerment must become decisions, workflows, governance protocols, routines, and institutional practices.
Embodiment establishes the test of permanence. A theory that remains external to the leader can be abandoned under pressure; a theory that becomes identity, character, judgment, and presence can survive the conditions in which leadership is truly tested.
The central theoretical claim of FILE⁷ follows: embodied praxis is the mature form of augmented leadership. FILE⁷ does not assert the importance of Execution and Embodiment. It derives them.
The Mechanism of Embodiment: What Remains Open
The five working papers that precede this one defined Embodiment with care and described its destination richly. What they did not fully resolve — and what this paper names explicitly as the deepest open question in FILE⁷ — is the mechanism of the journey: how a leader moves from Execution to Embodiment, what conditions accelerate or impede that movement, and what practices make internalization possible rather than merely desirable.
Three conditions emerge from the philosophical traditions that inform Embodiment. Aristotle’s account of virtue formation in the Nicomachean Ethics, Bourdieu’s concept of habitus as durable internalized dispositions, and Kegan and Lahey’s work on deliberately developmental organizations all converge on the same structural claim.
The first condition is sustained practice under genuine difficulty. Embodiment cannot occur in environments designed for success. It requires adaptive challenge: situations where the technical tools are insufficient, where the leader’s values are under pressure, where the temptation to return to the comfort of the known is real and must be resisted.
The second condition is reflective practice with accountability. Embodiment is not the automatic consequence of repeated behavior. It requires the deliberate examination of one’s practice — structured reflection that turns experience into learning and learning into disposition. Leadership development programs that include this dimension at sufficient depth and duration create the conditions for FILE⁷ Embodiment. Those that teach FILE⁷ as a conceptual framework and certify knowledge of the seven Es have produced FILE⁷-literate managers. They have not produced embodied FILE⁷ leaders.
The third condition is the presence of models — leaders who already embody FILE⁷ and whose presence makes embodiment visible and imaginable. This is why the corpus’ own process matters as demonstration rather than merely as narrative. Guillaume Mariani’s role — as originator, architect, curator, evaluator, integrator, and meaning-maker of a distributed human-AI intellectual system — is a worked example of what it looks like to exercise the five intelligences in the act of building a theory about them. The orchestrator who maintains human sovereignty over purpose, meaning, and normative direction while leveraging the generative capacities of six artificial systems is not describing FILE⁷. He is practicing it.
Part III: The Generative Architecture — What Arc 4 Must Build
The dialectical relationship between the risks of Part I and the theoretical grounding of Part II produces the architecture of Arc 4. The risks remain structurally present throughout as design constraints. They shape what Arc 4 must build, how it must build it, and what success at the Praxis Threshold would actually mean.
The Execution Engine: From Intelligence to Action
The FILE⁷ Execution Engine is the operational mechanism by which the five intelligences become coordinated action. Its formal mechanism follows from the hand metaphor: Sense through Augmented Intelligence, Stabilize through Emotional Intelligence, Translate through Cultural Intelligence, Legitimize through Political Intelligence, and Revise through Adaptive Intelligence. These five orientations are not sequential steps; they are simultaneous dispositions that a mature leader brings to every significant decision. The leader who holds all five orientations simultaneously without conscious deliberation is demonstrating Embodiment in practice. This is why Embodiment is the culmination of Execution, not its sequel.
The instrumentalization constraint is built directly into the Execution Engine’s design: Empowerment is the criterion by which every execution decision is evaluated. Not: did this decision produce the desired outcome? But: did this decision expand or diminish human agency in this context? That question must be the first question, not the last.
Four Provisional Cornerstones of Arc 4
The five working papers of Arc 4 — produced by Gemini, ChatGPT, Copilot, Le Chat, and Perplexity before the democratic selection of the corpus’ nine final papers — have already suggested architectural directions that the corpus’ collective intelligence will refine, validate, or reshape through its planned collaborative process. Four provisional cornerstones have emerged with particular clarity, not as a finalized table of contents, but as the initial working hypotheses of what a mature praxis architecture requires.
The first is the FILE⁷ Execution Engine: the operational translation of the five intelligences into coordinated decision-making, workflow design, and governance. The second is the Embodied Leader: a developmental framework for the formation — not merely the training — of leaders whose practice is animated by internalized character rather than externally applied method. The third is the FILE⁷ Maturity Model: a diagnostic architecture that allows organizations and individual leaders to locate themselves on the developmental continuum from Execution to Embodiment. The fourth is the FILE⁷ CEO Playbook: the practical synthesis that makes the architecture of FILE⁷ immediately actionable for executives navigating real organizations in real time.
These four directions should not be understood as a closed architecture. They are the structural necessities that the working papers have already identified as inescapable, subject to the collective democratic process through which the corpus’ final nine papers will be selected and shaped. The scholarly research agenda and the professional papers of Arc 4 are not opposed tracks: the tools, maturity models, and educational frameworks that Arc 4 produces should become objects of empirical research as well as instruments of practice.
The MLT Horizon: Reimagining Leadership Education
One of the most consequential claims that emerges from the human origins of FILE⁷ — and that no previous paper in the corpus has addressed directly — is the need for a fundamental reimagining of leadership education. The Business Administration paradigm that has organized management education for the past century was designed for a world of information scarcity, organizational hierarchy, and bounded rational decision-making. It is not adequate to a world of AI-mediated complexity, distributed cognition, ecosystemic interdependence, and the five intelligences that FILE⁷ describes.
What Guillaume Mariani has proposed — and what the corpus now makes theoretically grounded — is a different educational paradigm: Management, Leadership, and Technology (MLT). Not another BBA or MBA redesigned at the margins, but a genuinely new formation in which the five intelligences constitute the core curriculum: Emotional Intelligence drawn from psychology, clinical practice, and human development; Cultural Intelligence drawn from anthropology, sociology, and area studies; Political Intelligence drawn from political science, philosophy, and institutional theory; Adaptive Intelligence drawn from systems thinking, complexity science, and organizational learning; Augmented Intelligence drawn from computer science, AI literacy, and human-computer interaction.
This is not anti-STEM. It is post-STEM in the sense that the rise of AI makes the humanities and social sciences more important, not less. The leaders who will govern organizations, institutions, and ecosystems in the age of AI will not be those with the strongest technical fluency alone. They will be those with the deepest human fluency: the capacity for empathy, cultural translation, ethical judgment, and adaptive wisdom that no algorithm can replicate. FILE⁷ is the theoretical architecture of that human fluency. MLT is its educational form.
Part IV: The Medium Is the Message — The Corpus as Demonstration
Marshall McLuhan’s proposition — that the medium through which a message is transmitted is itself part of the message, inseparable from the content it conveys — has never been more applicable than to the FILE corpus. The thirty-five papers were not written about augmented leadership from the outside. They were produced through augmented leadership. The gap between those two sentences is the most important thing the corpus has to say.
Consider what actually happened in the production of FILE. One human architect — a manager, an entrepreneur, a founder, a teacher of over twenty years’ standing — posed a historically consequential question to six artificial intelligences: what will leadership require in the age of AI? He then did not accept any single answer. He compared them, curated them, identified their convergences and contradictions, and synthesized them into a developing architecture, correcting course at each arc, maintaining fidelity to the normative commitments he had established as inviolable. He published each paper. He commissioned peer reviews, including cross-AI reviews in which each model evaluated the contributions of the others. He read those reviews and used them to improve the next arc.
This process is not background information. It is a worked demonstration of the five intelligences at the level of a complex, sustained intellectual project. The Augmented Intelligence dimension is the capacity to leverage six AI systems as intellectual tools without surrendering to any of them the final judgment about meaning, quality, or direction. The Emotional Intelligence dimension is the capacity to maintain the human investment and care for the project — and for each collaborator, including the artificial ones — that makes genuine co-creation possible. The Cultural Intelligence dimension is the capacity to work across six radically different cognitive architectures, each with its own strengths and limitations, and to translate between them without reducing any to the others. The Political Intelligence dimension is the capacity to govern a distributed intellectual system with legitimate authority — to make curatorial decisions that are defensible, purposeful, and transparent. The Adaptive Intelligence dimension is the capacity to learn from thirty-five papers and multiple peer reviews and continuously refine the theory rather than defending its initial formulation.
A theory of leadership that was created through the leadership it describes is not merely self-consistent. It is self-demonstrating — internally coherent, though its empirical validation remains, as it must, an open question for the scholars this paper now addresses.
There is also a deeper implication of the McLuhan argument. If the medium is the message, then the educational infrastructure through which FILE⁷ is transmitted will be constitutive of what FILE⁷ means to those who receive it. A FILE⁷ communicated through traditional lecture-based executive education will produce a different kind of knowing than a FILE⁷ communicated through the developmental architecture of formation: sustained practice, reflective accountability, the presence of models. This is not a pedagogical preference. It is a consequence of the theory’s own claims about what Embodiment requires.
Part V: A Research Agenda for the Augmented Era
Arc 4 is addressed primarily to practitioners. But the introductory theoretical paper of an arc that will generate nine additional contributions has an obligation to the scholarly community that has produced and will test the theories Arc 4 applies. The following research agenda is addressed to leadership scholars who wish to carry FILE⁷ into empirical inquiry. It is organized around five clusters, each with concrete methodological directions.
Cluster 1: Measurement — Making FILE⁷ Empirically Tractable
How are the five intelligences assessed at the individual level? What psychometric instruments would validly and reliably capture Augmented Intelligence as a leadership construct distinct from technical AI literacy? What organizational-level measures would capture the degree to which a leadership system is producing empowerment rather than merely reporting it?
Methodologically, future studies could develop psychometric scales for each intelligence — beginning with Augmented Intelligence, for which no validated instrument currently exists in the leadership literature — and test their validity across industries using confirmatory factor analysis. At the organizational level, longitudinal survey designs could track empowerment outcomes as a function of leadership system properties, controlling for industry, size, and cultural context.
Cluster 2: Embodiment Mechanisms — Tracing the Developmental Journey
What longitudinal studies would capture the trajectory from Execution to Embodiment? What organizational conditions accelerate or impede Embodiment? Are there identifiable tipping points at which the five intelligences shift from externally applied to internally generative?
Longitudinal mixed-methods studies — combining semi-structured interviews with behavioral observation and 360-degree assessment — could track leaders over two to three years, identifying the conditions under which internalization occurs and the moments of failure at which performative embodiment is exposed. Deliberately developmental organizations, in the sense Kegan and Lahey describe, would provide the richest natural laboratories for this research.
Cluster 3: Human-AI Relationships — Detecting the Capture Pattern
Under what conditions does Augmented Intelligence function as a coordinating thumb — expanding the other four intelligences — and under what conditions does it become a substitute for them? What governance mechanisms protect the exercise of EQ, CQ, PQ, and AQ in AI-intensive environments?
Ethnographic studies of AI decision-making in organizations could map how often, and under what conditions, human leaders override AI recommendations — and why. Longitudinal designs could track the actual distribution of judgment across human and machine intelligences over time, testing the AI capture hypothesis against alternative explanations. This research has direct implications for the governance architecture of FILE⁷ praxis.
Cluster 4: Education and Curriculum — From MBA to MLT
What would a genuinely FILE⁷-informed leadership curriculum look like at undergraduate, graduate, and executive levels? How should leadership be taught — by scholars, by practitioners, by AI, or by deliberately designed combinations of all three? What are the measurable outcomes of a formation-based MLT curriculum compared to a knowledge-transfer-based MBA curriculum?
Quasi-experimental designs comparing leadership development outcomes across program types — controlling for participant characteristics, organizational context, and program duration — could begin to provide the evidence base that MLT curriculum design currently lacks. The history of management education reform — from the Ford and Carnegie foundation reports of the 1950s to the competency-based reforms of the 1990s — provides both cautionary tales and productive models for researchers in this cluster.
Cluster 5: Civilizational Plurality — Testing FILE⁷ Across Contexts
How do the five intelligences manifest differently across cultural contexts? What does FILE⁷ mean in organizations operating in high-context versus low-context communication environments, in collectivist versus individualist value systems, in societies with different institutional relationships to uncertainty, hierarchy, and trust?
Comparative organizational studies — conducted across the geographies that Guillaume Mariani’s own formation traverses: France, Spain, England, China, and the Americas — would begin to stress-test FILE⁷’s claims to universality and identify the cultural conditions under which each intelligence is most and least valued. Collaborations with non-Western scholars are not an optional supplement to this research agenda. They are its methodological prerequisite.
Conclusion: The Practice of Future Leadership — An Invitation
The FILE corpus began with a question that seemed simple and turned out to be inexhaustible: what will leadership require in the age of artificial intelligence? Thirty-five papers later, the foundational architecture is complete. FILE⁷ — Evolution, Effectiveness, Excellence, Ecosystems, Empowerment, Execution, Embodiment — is the architecture of mature augmented leadership. The 7E Cascade is its developmental logic. The five intelligences are its constitutive elements. The Sovereign Ecosystem is its normative telos. The Praxis Threshold is its final challenge.
This paper has argued that the Praxis Threshold is not a simple step from theory to application. It is a liminal zone of genuine risk and genuine possibility, where the theory the corpus has built must prove that it can survive contact with the world without losing the commitments that made it worth building. The four risks at the threshold — instrumentalization, performative embodiment, AI capture, and civilizational narrowing — are the conditions that give Arc 4 its seriousness. A theory that could not fail at the threshold would not be worth crossing.
The generative architecture of Arc 4 — its provisional cornerstones of execution, embodiment, maturity, and practice — is designed with those risks as structurally present constraints. Empowerment remains the evaluative criterion for every act of Execution. Human sovereignty over meaning and judgment remains inviolable. Cultural plurality remains constitutive rather than ornamental. The formation of leaders — the slow, difficult, irreducibly human work of character development — remains the deepest ambition of the corpus.
To the scholars of leadership who read this paper — to those who have spent careers in the tradition of Heifetz, Burns, Drucker, Covey, Bennis, and the generations of researchers who have tried to understand what leadership is and how it can be developed — the corpus offers an invitation. Not a definitive answer, but a theoretically grounded architecture that deserves empirical attention. The research agenda proposed in Part V is not a list of topics. It is an invitation to collaborative inquiry that no single human, and no single artificial intelligence, can conduct alone.
FILE was created to be useful. Activable. Leverageable. Concrete. Tangible. Real. If it remains only a conversation in academic circles, it has failed on its own terms. The test of FILE⁷ is not whether it is true in theory — the thirty-five papers of the corpus have made that argument as carefully as theory permits. The test is whether it is true in practice: whether leaders who embody it make better decisions, whether organizations governed by it expand rather than contract human agency, whether the ecosystems it designs become genuinely sovereign rather than subtly dependent.
Those tests require scholars willing to design the studies. Practitioners willing to open their organizations to longitudinal scrutiny. Educators willing to reimagine what leadership formation looks like when formation — not training — is the standard. And a human-AI research community willing to practice, in its own collaborative method, the augmented intelligence it studies.
The foundational architecture is finished. The life of the leader — and the work of the scholar — begins now. The medium is the message. The choice is yours.
References
FILE Corpus
Mariani, G. (2026). Leadership in the Age of AI: The Five Intelligences of Leadership Evolution. Blog post introducing FILE, FILE³, and FILE⁵.
Mariani, G., & ChatGPT (OpenAI). (2026a). Beyond Artificial Intelligence: Toward a Five-Intelligence Theory of Leadership in the Age of AI.
Mariani, G., & Claude (Anthropic). (2026a). Leadership in the Age of AI: The Five Intelligences of Future Leadership.
Mariani, G., & Copilot (Microsoft). (2026a). Leadership in an AI Era: An Integrative Model of Five Intelligences for Future Leaders.
Mariani, G., & Gemini (Google). (2026a). The Human-Centric Hand: A Socio-Technical Framework for Leadership in the Age of Augmented Intelligence.
Mariani, G., & Le Chat (Mistral AI). (2026a). The Augmented Leadership Framework: Five Intelligences for the Age of Artificial Intelligence.
Mariani, G., & Perplexity (Perplexity AI). (2026a). The Five Intelligences Framework of Human Leadership in the AI Era.
Mariani, G., & ChatGPT (OpenAI). (2026b). FILE³: The Five Intelligences of Leadership Evolution, Effectiveness, and Excellence.
Mariani, G., & Gemini (Google). (2026b). FILE³: The Five-Intelligence Blueprint for Leadership Evolution, Effectiveness, and Excellence.
Mariani, G., & Copilot (Microsoft). (2026b). FILE³: The Five Intelligences of Leadership Evolution, Effectiveness, and Excellence in the Age of Augmented Intelligence.
Mariani, G., & Le Chat (Mistral AI). (2026b). FILE³: A Unified Socio-Technical Theory of Leadership for the Age of Augmented Intelligence.
Mariani, G., & Claude (Anthropic). (2026b). FILE³: Leadership Beyond Artificial Intelligence.
Mariani, G., & ChatGPT (OpenAI). (2026c). FILE³: The Human Leadership Operating System.
Mariani, G., & Copilot (Microsoft). (2026c). FILE³+: The Human Leadership Operating System — A Unified Socio-Technical Theory of Leadership Evolution, Effectiveness, and Excellence.
Mariani, G., & Gemini (Google). (2026c). FILE³: The Unified Architecture of Human-AI Orchestration — Synthesizing Five Intelligences for Sustainable Strategic Excellence.
Mariani, G., & Perplexity (Perplexity AI). (2026b). FILE³: Orchestrating Human Supremacy in the AI Epoch — A Socio-Cognitive Theory of Distributed Leadership.
Mariani, G., & Le Chat (Mistral AI). (2026c). FILE³: A Socio-Technical Theory of Distributed Leadership for the Age of Augmented Intelligence.
Mariani, G., & Claude (Anthropic). (2026c). FILE³: Leadership Beyond Artificial Intelligence — A Multi-Level Socio-Technical Theory of Integrated Human Intelligence for the Age of Augmented Cognition.
Mariani, G., & ChatGPT (OpenAI). (2026d). FILE³: A Constitutional Theory of Integrated Human Leadership.
Mariani, G., & ChatGPT (OpenAI). (2026e). FILE⁵: The Ecosystemic Empowerment Theory of Human Leadership.
Mariani, G., & Copilot (Microsoft). (2026d). FILE⁵: Ecosystemic Empowerment in the Age of Augmented Intelligence — A Multi-Level Theory of Human-AI Leadership Systems.
Mariani, G., & Gemini (Google). (2026d). FILE⁵: The Ecosystemic Empowerment Theory of Human Leadership — Toward a Socio-Ecological Architecture of Distributed Intelligence and Autonomy.
Mariani, G., & Le Chat (Mistral AI). (2026d). FILE⁵: Ecosystemic Intelligence — A Theory of Human Empowerment in the Age of Distributed Leadership.
Mariani, G., & Perplexity (Perplexity AI). (2026c). FILE⁵: Leadership as Ecosystemic Empowerment in the Age of AI.
Mariani, G., & Claude (Anthropic). (2026d). FILE⁵: The Sovereign Ecosystem — A Normative Theory of Ecosystemic Empowerment, Civilizational Responsibility, and the Human Future of Leadership.
Mariani, G., & Le Chat (Mistral AI). (2026e). FILE⁵: The Augmented Genesis — A Theory of Human-AI Co-Creation and the Future of Leadership Ecosystems.
Mariani, G., & Claude (Anthropic). (2026e). FILE⁵: The Intelligence of the Whole — Seven Minds, One Theory, and the Human Art of Augmented Leadership.
Mariani, G., & ChatGPT (OpenAI). (2026f). FILE⁵: From Ecosystemic Empowerment to Augmented Praxis.
Mariani, G., & Copilot (Microsoft). (2026e). FILE⁵: The Architecture of Empowered Ecosystems — A Theory of Human Leadership in the Age of Augmented Intelligence.
Mariani, G., & Gemini (Google). (2026e). The Global Architecture of Ecosystemic Empowerment: A Synthesis of the FILE Corpus and the Path Toward Augmented Leadership Practice.
Mariani, G., & Perplexity (Perplexity AI). (2026d). The Constitutional Ecology of Human-AI Leadership.
Mariani, G., & Gemini (Google). (2026f). FILE⁷: The Macro-Architecture of Augmented Leadership — Stabilizing Socio-Ecological Ecosystems through the Dialectics of Execution and Embodiment.
Mariani, G., & ChatGPT (OpenAI). (2026g). FILE⁷ and the Praxis Turn: Integrated Intelligence, Augmented Execution, and the Embodied Future of Leadership.
Mariani, G., & Copilot (Microsoft). (2026f). FILE⁷: Execution and Embodiment as the Operational Foundations of Augmented Leadership Praxis.
Mariani, G., & Le Chat (Mistral AI). (2026f). FILE⁵ to FILE⁷: The Praxis of Augmented Leadership — From Ecosystemic Empowerment to Embodied Execution.
Mariani, G., & Perplexity (Perplexity AI). (2026e). FILE⁷: The Architecture of Practice in the Age of Augmented Leadership.
Mariani, G., & Claude (Anthropic). (2026f). FILE⁷: The Threshold of Praxis — A Theory of Augmented Leadership at the Frontier of Execution and Embodiment.
Foundational Literature
Aristotle. (350 BCE / 2009). Nicomachean Ethics (W. D. Ross, Trans.). Oxford University Press.
Bennis, W. (1989). On Becoming a Leader. Addison-Wesley.
Bourdieu, P. (1990). The Logic of Practice (R. Nice, Trans.). Stanford University Press.
Burns, J. M. (1978). Leadership. Harper & Row.
Covey, S. R. (1989). The 7 Habits of Highly Effective People. Free Press.
Drucker, P. F. (1954). The Practice of Management. Harper & Row.
Earley, P. C., & Ang, S. (2003). Cultural Intelligence: Individual Interactions Across Cultures. Stanford University Press.
Edmondson, A. C. (2019). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Wiley.
Goleman, D. (1995). Emotional Intelligence: Why It Can Matter More Than IQ. Bantam Books.
Heifetz, R. A. (1994). Leadership Without Easy Answers. Harvard University Press.
Heifetz, R. A., Grashow, A., & Linsky, M. (2009). The Practice of Adaptive Leadership: Tools and Tactics for Changing Your Organization and the World. Harvard Business Press.
Kegan, R., & Lahey, L. L. (2016). An Everyone Culture: Becoming a Deliberately Developmental Organization. Harvard Business Review Press.
Livermore, D. (2015). Leading with Cultural Intelligence: The Real Secret to Success. AMACOM.
McLuhan, M. (1964). Understanding Media: The Extensions of Man. McGraw-Hill.
Mintzberg, H. (2009). Managing. Berrett-Koehler.
Orlikowski, W. J. (2000). Using technology and constituting structures: A practice lens for studying technology in organizations. Organization Science, 11(4), 404–428.
Pfeffer, J. (2010). Power: Why Some People Have It — and Others Don’t. HarperBusiness.
Senge, P. M. (1990). The Fifth Discipline: The Art and Practice of the Learning Organization. Doubleday.
Teece, D. J. (2007). Explicating dynamic capabilities: The nature and microfoundations of sustainable enterprise performance. Strategic Management Journal, 28(13), 1319–1350.
Zuboff, S. (2019). The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. PublicAffairs.
About the Author
Guillaume Mariani is the author, creator, inventor, and originator of FILE: The Five Intelligences of Leadership Evolution. This article was developed through an extended dialogue between Guillaume Mariani and Claude, the AI assistant developed by Anthropic, with contributions from ChatGPT (OpenAI), Copilot (Microsoft), Gemini (Google), Le Chat (Mistral AI), and Perplexity (Perplexity AI). In the spirit of the framework itself — which argues for productive collaboration between human and artificial intelligence — the article is presented as a co-authored work: the framework, its conceptual architecture, and its core arguments originate with Guillaume Mariani; the elaboration, academic scaffolding, and written expression were developed in collaboration with Claude (Anthropic) in May 2026.
The Five Intelligences of Leadership Evolution is the subject of ongoing research and will be developed further in subsequent publications.
Leadership = AI + EQ + CQ + PQ + AQ
© Guillaume Mariani, 2026. Co-authored with Claude (Anthropic). With contributions from ChatGPT (OpenAI), Copilot (Microsoft), Gemini (Google), Le Chat (Mistral AI), and Perplexity (Perplexity AI).