Leading Across the Spectrum of Human-AI Relationships: A Conceptual Framework for Increasingly Heterogeneous Teams
Pith reviewed 2026-05-07 09:26 UTC · model grok-4.3
The pith
A five-position spectrum helps leaders recognize when AI has taken over decision authority even as humans remain visible.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
The paper claims that consequential decisions are shaped by one of five landmark configurations along a spectrum from pure human to pure AI, defined by where problem framing, work redirection, and responsibility for outcomes actually occur. These positions serve as reference points that leaders can use to detect when a setup has layered, drifted, or changed, even within one decision cycle. The central mechanism is co-adaptability, the capacity for human and non-human participants in heterogeneous teams to adjust together and improve the configuration over time.
What carries the argument
The five-position spectrum (Pure Human, Centaur, Co-equal, Minotaur, Pure AI) that marks the location of leadership work in framing problems, redirecting effort, and answering for results.
If this is right
- Leaders can detect when a configuration has shifted and judge whether it still suits the decision at hand.
- Misrecognition produces either ceremonial human oversight or human involvement that worsens results.
- Power, responsibility, and trust in organizations follow the actual configuration rather than the declared one.
- Heterogeneous teams differ in number, substrate, model architecture, capability, speed, memory, and participation form.
- The governability of future decisions depends on early detection of where shaping authority resides.
Where Pith is reading between the lines
- The spectrum could be mapped onto real organizational cases to measure how often leaders correctly locate decision authority before and after training on the framework.
- It offers a way to revisit accountability questions in AI ethics by focusing on recognition of actual influence rather than formal roles.
- Extensions might include simple checklists leaders could use during meetings to note the current configuration and flag potential drifts.
Load-bearing premise
The spectrum and co-adaptability idea give leaders enough practical visibility to identify and adjust configurations without needing further empirical checks in specific settings.
What would settle it
A controlled comparison in which leaders apply the spectrum to identify who actually framed, redirected, and answered for outcomes in mixed-team scenarios versus leaders without the spectrum, measuring differences in accuracy of recognition.
read the original abstract
What shapes a consequential decision when human and artificial intelligence work on it together? The answer is becoming harder to see. A decision may look human-led after AI has set the frame, or appear automated while human judgment still carries decisive force. This paper offers a leadership-facing spectrum to see those relationships within a bounded mandate: Pure Human, Centaur (human-dominant, with AI in the loop), Co-equal, Minotaur (AI-dominant, with humans in the loop), and Pure AI. The spectrum asks where leadership work occurs: who frames the problem, who redirects the work, and who can answer for what follows. The five positions are landmarks that help leaders recognize configurations as they layer, drift, or change in a single decision. The central risk is misrecognition: leaders may keep a human-centered story in place after decision-shaping authority has shifted elsewhere. They may believe oversight remains meaningful when it has become ceremonial, or keep humans in the loop when their involvement could make the decision worse. The framework introduces co-adaptability, the capacity of a configuration to improve as human and non-human participants adjust together, and places it within heterogeneous teaming, where participants may vary by number, substrate, model architecture, capability, speed, memory, and form of participation. The aim is practical: to help strategic leaders and those designing or deploying AI systems recognize the configuration at work, notice when it shifts, and judge whether it fits the decision before them. These configurations will shape how power, responsibility, and trust are distributed in organizational life. Whether the futures they help create remain governable and worth inhabiting will depend on leaders who can see, early enough, where and how consequential decisions are actually being shaped.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript proposes a conceptual framework for analyzing human-AI decision-making in heterogeneous teams. It defines a five-position spectrum (Pure Human, Centaur, Co-equal, Minotaur, Pure AI) based on the location of leadership work—problem framing, work redirection, and accountability for outcomes. The framework introduces co-adaptability as the capacity of such configurations to improve through mutual adjustment among participants that may differ in number, substrate, architecture, capability, speed, memory, and participation form. The central claim is that leaders can use these landmarks to detect shifts in decision-shaping authority, avoid misrecognition (e.g., treating ceremonial oversight as meaningful), and assess fit for a given decision.
Significance. If the distinctions hold, the framework offers a coherent classification tool that could help surface otherwise invisible reallocations of authority in AI-augmented organizations. It receives credit for maintaining internal consistency without hidden premises about measurability and for framing co-adaptability as a dynamic property rather than a static state. Its potential significance lies in influencing how responsibility and trust are conceptualized in mixed teams, though this remains prospective given the absence of tested applications.
major comments (2)
- [Abstract] Abstract and the section defining the spectrum: the claim that the five positions function as 'landmarks that help leaders recognize configurations as they layer, drift, or change' and enable judgment of fit is load-bearing for the stated practical aim, yet the manuscript provides no criteria, indicators, or decision procedure for identifying a position or assessing fit in a concrete case.
- [co-adaptability discussion] The section introducing co-adaptability: the assertion that this capacity allows configurations to 'improve as human and non-human participants adjust together' is presented without distinguishing it from existing concepts in adaptive teaming or human-AI collaboration literature, leaving open whether it adds a distinct, non-redundant lens.
minor comments (2)
- The metaphor 'Minotaur' for the AI-dominant position is introduced without a brief gloss on its intended meaning, which could reduce clarity for readers unfamiliar with the classical reference.
- The manuscript would benefit from explicit pointers to related work on human-AI teaming taxonomies or leadership in sociotechnical systems to situate the novelty of the spectrum.
Simulated Author's Rebuttal
We thank the referee for the constructive review and for recognizing the framework's internal consistency and prospective value. We address each major comment below with targeted revisions to improve clarity on application and novelty without altering the conceptual focus.
read point-by-point responses
-
Referee: [Abstract] Abstract and the section defining the spectrum: the claim that the five positions function as 'landmarks that help leaders recognize configurations as they layer, drift, or change' and enable judgment of fit is load-bearing for the stated practical aim, yet the manuscript provides no criteria, indicators, or decision procedure for identifying a position or assessing fit in a concrete case.
Authors: We accept that the manuscript lacks explicit criteria, indicators, or a formal decision procedure, as the framework is designed as an interpretive conceptual tool rather than a diagnostic checklist. To strengthen the practical aim, we will add a new subsection with three brief hypothetical scenarios illustrating qualitative indicators (e.g., who authors the initial problem statement, where redirection occurs mid-process, and who bears accountability). We will also add an explicit limitations paragraph noting that classification relies on leader judgment and is not intended for automation or quantification. This addresses the concern while preserving the framework's scope. revision: partial
-
Referee: [co-adaptability discussion] The section introducing co-adaptability: the assertion that this capacity allows configurations to 'improve as human and non-human participants adjust together' is presented without distinguishing it from existing concepts in adaptive teaming or human-AI collaboration literature, leaving open whether it adds a distinct, non-redundant lens.
Authors: We agree that explicit differentiation is needed to establish non-redundancy. Co-adaptability is defined as mutual adjustment that can itself shift the leadership position across substrate and architectural differences. We will revise the section to include a short comparison: unlike adaptive teaming (typically intra-human) or standard human-AI collaboration (often static role assignment), co-adaptability emphasizes dynamic reconfiguration where adjustment may move authority between human and AI participants. This addition will be made in the next version. revision: yes
Circularity Check
No significant circularity detected
full rationale
The paper advances a purely conceptual framework with a five-position spectrum (Pure Human, Centaur, Co-equal, Minotaur, Pure AI) and the notion of co-adaptability for heterogeneous teams. It contains no equations, derivations, fitted parameters, quantitative predictions, or formal proofs. The central claim—that leaders can use these landmarks to detect shifts in decision-shaping authority—rests solely on the internal coherence and descriptive utility of the distinctions, without any reduction to self-definitional inputs, self-citation chains, or renamed prior results. No load-bearing step in the provided text exhibits circularity under the enumerated patterns.
Axiom & Free-Parameter Ledger
axioms (1)
- domain assumption Leadership in decisions occurs through framing the problem, redirecting the work, and answering for outcomes.
Reference graph
Works this paper leans on
-
[1]
Bevilacqua S, Ferraris A, Matzler K, Kuděj M. Strategic leadership at high altitude: Investigating how AI affects the required skills of top managers. J Bus Res. 2026 Feb 1;205:115878
work page 2026
-
[2]
Sposato M. Leading hybrid intelligence systems: a general systems theory approach to organizational leadership in hybrid intelligence knowledge ecosystems. VINE J Inf Knowl Manag Syst. 2026 Feb 9;1 –
work page 2026
-
[3]
Human agency in AI configurations supporting organizational decision - making
Wang BY, Boell SK, Riemer K, Peter S. Human agency in AI configurations supporting organizational decision - making. In: ACIS 2023 Proceedings Australasian (ACIS)
work page 2023
-
[4]
AI and organizational leadership: bibliometric review and future trends
González - Reyes C, Ficapal - Cusí P, Torrent - Sellens J. AI and organizational leadership: bibliometric review and future trends. J Organ Chang Manag. 2026 Feb 10;1 –
work page 2026
-
[5]
Li H, Tian F. Advancing decision - making through AI - human collaboration: A systematic review and conceptual framework. Group Decis Negot. 2026 Jun 3;35(2):26
work page 2026
-
[6]
Bevilacqua S, Masárová J, Perotti FA, Ferraris A. Enhancing top managers’ leadership with artificial intelligence: insights from a systematic literature review. Rev Manag Sci. 2025 Sep;19(9):2899 –
work page 2025
-
[7]
Available from: http://arxiv.org/abs/2406.10942
arXiv:2406.10942. Available from: http://arxiv.org/abs/2406.10942
-
[8]
Minotaurs, not centaurs: The future of manned - unmanned teaming
Sparrow RJ, Henschke A. Minotaurs, not centaurs: The future of manned - unmanned teaming. Parameters. 2023 Mar 3;53(1):14. 11 Figure. The spectrum of human - AI relationships The five positions are illustrative waypoints along a continuous spectrum. Real cases may occupy intermediate, layered, or mixed configurations. 12 Table. Practical diagnostic guide ...
work page 2023
-
[9]
The scholarly strategy was refined iteratively, beginning with broad combinations of leadership - centered and AI - centered terms and narrowing as results became dominated by clinically focus ed, governance - heavy, or otherwise adjacent literature. One documented Scopus strategy combined title terms such as “strategic leadership,” “organizational leader...
work page 2022
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.