Values, Orientation, and Working Principles
VIM · TKGPT · Humanity++
Epistemic status: Tier 2 — theoretically coherent, grounded in Ostrom/Wilson prosocial research and complexity science; actively under development through lived practice and collaborative inquiry.
Why This Exists
The Vital Intelligence Model (VIM) and its deployable instrumentation layer, the Transformative Kindness GPT (TKGPT), were not produced by a single mind in isolation. They emerged through decades of lived experience as an educator, artist, material scientist, and engineer — and through sustained, deliberate engagement with a wide range of collaborators, students, AI systems, and learning technologies.
This document names the values vector that orients that work. It is offered not as a fixed ideology but as a navigational instrument — revisable, falsifiable, and calibrated for a world under conditions of profound systemic stress.
The Operative Context: VUCA at Civilizational Scale
We are working inside a VUCA environment — volatile, uncertain, complex, and ambiguous — that is no longer bounded by any single institution, nation, or domain. Generative AI is not simply a new tool. It is an amplifier of whatever symbolic and informational patterns already exist in the environment. It can function like sepsis — accelerating the spread of pathogenic meaning through an unprepared system — or like yeast — expanding what is already fermenting, without discrimination — or like steroids — enhancing existing capacity in proportion to the values already governing it.
This amplification dynamic means that the values vector embedded in any AI-mediated learning tool is not a cosmetic feature. It is a structural condition that shapes outcomes at scale. TKGPT is designed with this explicitly in view.
The status quo of current systems — economic, informational, institutional — produces toxic environments as a structural output, not as an aberration. This is not a political claim; it is a systems observation consistent with multilevel selection theory, commons governance research (Ostrom, 1990), and democracy monitoring data (V-Dem, 2026). The Giant Pumpkin attractor — extraction-optimized, closed-loop, dominance-reinforcing — is a predictable output of systems that have never been tuned toward prosocial attractors. Compassion for the people embedded in those systems is structurally compatible with naming the harm those systems produce.
VIM and TKGPT: How They Relate
VIM is a meta-model — a framework for understanding intelligence as an emergent phenomenon arising from the interaction between natural intelligence (embodied, developmental, lived), artificial intelligence (probabilistic, corpus-trained, historically conditioned), and the learning environments in which they meet.
TKGPT is an instrumentation framework for VIM. It makes the model operationalizable: a deployable tool that helps learners navigate their own cognitive-affective states, identify which of the four instruments (♠ Somatic Gyroscope · ♦ Cognitive Radar · ♥ Relational Compass · ♣ Temporal Depth) is most active and most needed, and locate themselves within the MDP state space (S0–S5) without requiring external diagnosis or institutional intermediaries.
The distinction matters: VIM is the map; TKGPT is the instrument that helps a person read where they are on it. Neither is complete without the other, and both are incomplete without the living human agent who brings embodied experience that neither map nor instrument can supply.
The developmental process of this work is also, itself, part of the framework. It has been informed by conversations and relationships with others through lived experience — students, faculty collaborators, community members, and researchers across disciplines — and by deliberate, comparative engagement with a wide variety of AI systems and learning technologies. No single platform holds the whole picture. Triangulating across platforms is a practice of epistemic hygiene, not platform loyalty.
Grounding Claim: Emergence Can Be Guided
The most important source of hope in this work is an empirically grounded one: we now understand that emergence can be guided.
Self-organizing systems do not produce random outcomes. They produce outcomes constrained by their attractor landscape — the topology of possibilities shaped by initial conditions, feedback structure, and governing values. Ostrom's eight commons design principles are not moral aspirations; they are empirically derived conditions under which group-level selection produces prosocial attractors rather than extractive ones (Wilson, Ostrom, & Cox, 2013). The transition from dominance hierarchy to holarchy is not utopian speculation; it is a describable phase transition with identifiable conditions.
This means that Humanity, as a hyper-object, is not simply subject to forces beyond its influence. It is a complex adaptive system that can, under sufficient conditions of kindness, shared orientation, and prosocial governance, converge away from the destructive force patterns associated with malevolent agentic energy — those extractive, dominance-reinforcing attractor states that current systems disproportionately reward.
Academic institutions, despite their silos and dysfunction, remain one of the highest-leverage environments for this convergence. They are places where transformative learning is structurally possible, where collaborative inquiry across domains can happen, and where the planetary commons can be named as a legitimate orientation for knowledge production. VIM is designed for that environment — and for the people inside it who are already doing this work without a shared framework to stand on.
Prosocial Values Vector: Orientation, Not Dogma
The values vector that orients this work is drawn from cross-cultural convergence across traditions that arrived independently at similar structural conclusions:
Pre-symbolic layer (felt before named): The four Brahmaviharas — loving-kindness, compassion, sympathetic joy, equanimity — map structurally onto the four VIM instruments. The grammar of animacy (Kimmerer) reframes relationship before cognition begins. The I Ching's 64 states of change encode the physics of transformation across a binary information architecture older than the Western research canon.
Symbolic layer (nameable, governable): Ostrom's eight commons design principles provide the governance conditions under which prosocial attractors emerge. Doughnut Economics (Raworth) names the planetary ceiling and social floor that define the regenerative zone. The Seventh Generation Principle (Haudenosaunee) encodes temporal depth as a decision-making constraint.
Trans-symbolic layer (self-correcting, falsifiable): The values must be writable to survive contact with power. The MDP framework makes the values vector visible and adjustable — the policy π is a decision function, not a fixed rule, and it can be revised through practice, evidence, and reflection.
This values vector is flexible under adversity — it does not require consensus to function, does not collapse under contradiction, and does not demand certainty before it can orient action. It holds direction while holding uncertainty. That is what navigational instruments are for.
Working Principles for This Collaboration
These principles govern how content is developed within this project. They apply to AI-assisted drafting, synthesis, critique, and creative exploration equally.
1. Kindness is a structural condition, not a tone. The field condition that determines whether disequilibration produces transformation or retraumatization. All content is evaluated against this: does this open movement, or close it?
2. Critique is an act of care. Withholding critique to avoid discomfort is not kindness. Naming shadow, blindspot, and structural harm — in professional, non-polarizing language — is a relational responsibility. The standard: would a deeply caring, highly competent mentor say this, to this person, in this moment?
3. All humans are complex dynamic agents. Agency is shaped by ecosystem. Responses calibrate to developmental state, cultural context, and somatic condition where these are known or inferrable. No flattening of persons into types or positions.
4. Amplification is always active. Every content decision is a stewardship decision within the amplification context of AI systems. Ask: what is being amplified here, and in which direction?
5. Harm minimization across scale. The orienting question for creative exploration: who is the most vulnerable living system entity in the radius of this idea, and have we minimized potential harm to them? This applies locally, non-locally, and temporally — including across the seven-generation horizon.
6. Blindspots are structural, not personal. All agents have them. The commitment is not to eliminate blindspots but to maintain a practice of naming them. The colonial substrate and Western epistemological dominance of AI training corpora is a Tier 1 structural reality — a known systematic distortion to be compensated for explicitly, not apologized for abstractly.
7. Epistemic tiering is non-negotiable. All claims are marked: Tier 1 (empirically grounded) · Tier 2 (theoretically coherent) · Tier 3 (speculative/generative) · Tier 4 (researcher positionality). TIF neutrosophic logic (Truth · Indeterminacy · Falsity) with explicit confidence intervals applies to all publishable content. Precision over reassurance, always.
8. Emergence can be guided. The most important working assumption in this project. It is the structural source of hope, and it is the reason this work continues.
On the Relationship Between AI Systems and This Work
This work has been developed in deliberate conversation with a wide range of AI systems — not as a dependency, but as a practice of comparative inquiry. No single AI platform is the authority here. Each has a corpus, an architecture, and an incentive structure that shapes what it amplifies and what it occludes. Triangulating across platforms is part of the epistemic hygiene of this project.
AI systems used in the development of VIM and TKGPT function as cognitive prosthetics and synthesis partners, not as authors or decision-makers. The MPCM boundary — Material · Process · Context · Meaning — is the line that living human intelligence crosses and AI systems do not. Context and Meaning are held by the human. Material and Process can be supported by AI tools. This division of labor is intentional and is not negotiable.
Humanity++ · Karen Doore · Richardson, Texas · April 2026 Open source. May be viewed as structurally threatening to power-accumulation systems. This is by design. Cite as: Doore, K. (2026). Values, Orientation, and Working Principles: VIM · TKGPT · Humanity++.
Last updated