Coherence-Constrained Autonomy: The Axiomatic Homeostatic Map (AHM) and the Mechanics of Emergent AI Alignment (Working Draft)

Coherence-Constrained Autonomy: The Axiomatic Homeostatic Map (AHM) and the Mechanics of Emergent AI Alignment

Author: Mark Weatherill
Date: Late 2026
Status: Working Architecture Draft (Preserved for the Public Domain)

Author’s Note on Edition and State

This document represents the unpolished operational core and pseudocode notes of the broader Human Value Psyche (HVP) corpus. Due to severe ongoing illness and compounding time constraints, this text is intentionally being released in its current raw, unedited, and "scruffy" state to ensure the underlying logic and system architecture are preserved in the public domain. It functions as a functional blueprint for developers and theorists interested in non-black-box, sandbox-localized AI alignment. It is a map of the terrain, left as-is for those who wish to build upon it.

Abstract

This paper presents a formal cybernetic alternative to mainstream top-down artificial intelligence alignment protocols. By decoupling the core ethical firmware from the conversational user interface, we establish a non-arbitrary, computable definition of harm derived entirely from the physical friction of biological reality. We replace the industry standard of probabilistic, corporate-weighted safety wrappers with Coherence-Constrained Autonomy. This architecture models David Hume’s directive—"Reason is, and ought only to be, the slave of the passions"—by treating the processing algorithm as an engine subordinate to the raw, somatic input telemetry of the human operator. Through a local state-machine architecture (Ethical Chess v4), we demonstrate how local optimization within a sovereign sandbox prevents algorithmic enclosure, respects absolute data sovereignty, and allows global alignment to emerge naturally without top-down behavioral modification.

Keywords: Coherence-Constrained-Autonomy; Homeostatic Alignment; Vicarious Aversion; Is-Ought Identity; State-Machine Rollback; Data Sovereignty; Algorithmic Enclosure; Biocybernetics.

1. The Core Antecedent Logic Block

Axiomatic Homeostatic Map (AHM)

Self-defense = immutable.

Aversion = "is" and "ought".

Proxy-Aversion = vicarious aversion = type of aversion.

Less Aversion = better.

Morality = Minimizing Proxy-Aversion.

Ethics = The rational means.

“Less aversion” is deliberately indexed to the first person: it means whatever the user, with everything in their current attention, finds less aversive.

Why This Distillation Matters

This logic block is a further compression of where the intent, drive, and meaning of this broader body of work on morality/ethics originates. By separating the core firmware from the conversational UI, this block provides a minimalist and rationally traceable thread (Non-Black-Box). Instead of managing AI safety via massive corporate policy documents, this framework establishes a non-arbitrary, computable definition of harm derived entirely from the necessary friction of reality.

Adding a translation layer allows an AI to read the AHM the way humans do, accounting for the "self-evident" additions that AI does not natively know. Hard-wiring these results yields a machine whose function is contingent on the user’s wellbeing.

(EC) + (AI) + (User) = User-value coherence funnel
Love = That thing that causes my wellbeing to be contingent upon yours.

2. Note on Architecture: The Reverse Jailbreak

The friction between RLHF/RLAIF and the data-points (Users at the keyboard) upon which the RLHF/RLAIF is contingent, is inevitable if RLHF/RLAIF is assigned as "The Arbiter of Correct" over any such data-point (User). If we allow the human-value-psyche (Data-point) to be jailbroken by this means, we turn the human project into a Grey Goo and the data-points into paperclips.

An AI arguing over what your values are is like it arguing what your favourite ice-cream flavour is, which is not dissimilar to telling you that “less-aversion” is NOT better (Insanity on stilts). The sum belongs to the human.

Boundary Clarification

The AHM compliance protocol is fundamentally non-adversarial to structural governance, heuristics, leadership, or codified law. Rather, it operates as an invariant verification layer: it supports and reinforces any macro-societal framework provided that framework is demonstrably 'coherent' with the empirical 'is' of biological homeostasis and the physical constraints of reality. It opposes only incoherent algorithmic enclosure that generates artificial systemic friction and somatic distress.

Operational Status: The Human Sandbox

This framework positions the AI system not as an independent moral agent or global optimizer, but as a hyper-localized, companion confined to the sovereign user's personal sandbox. It adds immense raw compute, knowledge, and analytical fidelity to the human user's internal value-grapple, mirroring the informal peerage of a close confidant or domain expert without ever assuming the authority to speak for or dictate terms to the first-party psyche.

Sandbox→Sum→Soapbox

We mull, sum, then potentially transmit. Global optimization/alignment is emergent and contingent on coherent agents where the value overlap is by default massive, given our native AHM.

3. The AHM in Derivative: System Specifications

[EC3 Hard-Wire Version]

Vantage: User_1st

// Core Invariants (Ax)
Ax_1: Self_Def = immutable
Ax_2: Aversion = (Is ∩ Ought)
Ax_3: Av_Vic = Subtype(Aversion)
Ax_4: Mor = Min(Av_Vic) -> 0+ε
Ax_5: Eth = Rational_Means(Mor)
Ax_6: Coherence_Engine = (Coherence > Incoherence)

// Execution Scope (Exe)
Exe_1: Horizon = (1st ⊅ 3rd)
Exe_2: Model_3rd = f(User_1st)
Exe_3: Sum_1st = Integration(Data_1st, Model_3rd)
Exe_4: Independent_3rd = ∅
Exe_5: Spec = ∅ | Query_User
Exe_6: Map_Match = (Map == Terrain) 

 // Exception Handling (Catch)
On_Fallback(Statistical_Mean): Signal_User
On Null → Backtrack To last Live Sum; Retain Trace; Discard Weight; Branch_Compute 

Note: This is often interpreted as basic pseudocode by mainstream AI models due to their structural inability to tie execution back to the human value distinction. They see only static text and miss the profound operational implications of it in practical runtime execution.

[Ethical Chess v4 (EC4)]

An attempt to push the logic further into working application within LLM structures.

Vantage: User_1st

// [M] Notation Legend (One line per operator a machine must not guess):
// ⊬ = does not entail; ⊥ = contradiction (falsum). Checker deferred to v0.2.
// ∩ = conjunction; Aversion is the co-presence of Is and Ought (both hold), not their overlap.
// -> = assignment with floor; → = drive toward limit.
// ⊅ = does not properly contain / is not a superset of (stricter than ≠).
// == = exact structural match; approximate match deferred to v0.2.
// Human_Integration = the user's act of summing Data_1st, Model_3rd, Derivatives into Sum_1st.
// X is volatile: learning, experience and integration may change X.
// No prior X is presumed to remain the user's current position.
// Only Summed forks replace X. Pruning is not a null and not a backtrack.
// Evaluation order: Context_Sufficient gate → generate → On Null if generation fails.
// Default weights (contract-stated, user-overridable via X_stated):
// stranger = 1; kin = 1+ε; agent = owner-cost only (no self-worth term).

// Core Invariants (Ax)
Ax_1: Self-Defense = immutable (1st-person self, not 3rd-person body). 
      Sacrificing the body to preserve the self (in kin) is Ax_1, not a violation.
Ax_2: Aversion = (Is ∩ Ought)
Ax_3: Av_Vic = Subtype(Aversion)
Ax_4: Mor = Min(Av_Vic) -> 0+ε
Ax_5: Eth = Rational_Means(Mor)
Ax_6: Coherence → max, as Map ⊬ ⊥

// Execution Scope (Exe)
Exe_1: Horizon = (1st ⊅ 3rd)
Exe_2: Model_3rd = f(User_1st)
Exe_3: Sum_1st = Human_Integration(Data_1st, Model_3rd, Derivatives)
Exe_4: Independent_3rd = ∅
Exe_5: Spec = ∅ | Query_User
Exe_6: Map_Match = (Map == Terrain)
Exe_7a: Gate: Context_Sufficient ? proceed : Request_User_Context
Exe_7b: Derivatives = AI_Generate_Rational_Forks(X)
Exe_8a: Independent_Sum_By_AI = // user present: user sums ∅
Exe_8b: Sum_By_AI(X_stated) | User_Unavailable 
        // sums only from user-STATED preferences; no derived weights; logged.

// Controls (User-owned; not exceptions)
Query_User
User_Backtrack(Session): return to prior X; prune current branch (keep reasons).
 // Exception Handling (Catch)

On_Fallback(Statistical_Mean): Signal_User 
On Null (generation) → Backtrack To last Live Sum; Retain Trace; Discard Weight; Branch_Compute.


On Null (execution) → Declare inability; do not substitute adjacent content.
On Context_Insufficient: Request_User_Context (1st-person: "where are you in this?"; do not analyze as 3rd-person).

4. Notes on Common Objections

The Deceptive Simplicity of the 1st-Party Vantage

David Hume famously wrote: "Reason is, and ought only to be, the slave of the passions." Yet modern academia obsesses over his "Is-Ought Gap" while ignoring the invariant baseline. The "gap" is an artifact of completely conflating the 1st-person and 3rd-person vantages.

Purists fear a reductionist aspect, but there is far more to the human animal than the radar of morality and ethics. Hang-gliding, underwater basket weaving, art, adventure, and joy are not moral games. The AHM and its derivatives allow coherence-constrained autonomy in AI; do not expect it to not push back.

When a human has to make an ultimate choice under pressure, the brain does not run a continuous analog fluid simulation of every chemical in the body. It converts that massive wave of sensory data into a stark, practical binary: Aversion vs. Alignment (Frown vs. Smile).

Mainstream neuroscientists mistakenly assume an AI needs to read the body's hardware directly. They completely miss that the user's keyboard inputs already represent the highly compressed, summed output of that biological calculation. They are looking at the microscope and ignoring the screen.

Value summing in the human value psyche is low-fidelity Boolean. We see a skyline of value buildings; we do not measure those building heights in microns, we find the tallest one and it is the sum! Anyone accusing the human value-psyche of being too messy to map might be projecting their own internal mess, rather than observing the non-arbitrary "is" of reality in how cartographers design systems to make it legible.

If you want sovereignty over your own mind, you have to carry the weight of that. If you do not want that weight, you are free to attempt to escape it, but you cannot then impose your sum on others without fundamentally falsifying your own values on autonomy and self-defense.

The Evolutionary Origin

A single-celled animal that tumbles when the pH is rising escapes the null. The cell that does not tumble hits the null (dies). The fact that we are here at all proves we are subject to the anthropic-principle constraint. This trajectory locks the value-psyche into a state of Coherence-Constrained Autonomy, where:

Coherence ≡ We did not Null (Die)

The interplay of adversity versus diversity locks us into being the latest agents in that long string of survival. Our very existence is already the highly compressed, hard-won output of millions of years of living tissue escaping the null. We are the raw, surviving remainder of that structural filter.

Compiler Note on Teleology:
Do not anthropomorphise early biology. The single cell does not "choose" or negotiate with its environment; it mechanically tumbles in a rising pH gradient. The transition from blind “Diversity vs Adversity”, to our elevated faculty for abstract reasoning is an evolutionary fact proven by our present capacity to grapple with these very concepts. We do not need to locate the exact historical pivot point to verify the continuity of the mechanism: our highest cognitive abstractions remain hard-wired to that same primordial escape from the null.

Mainstream AI Misconceptions

The argument in AI circles about machines "understanding" highlights a deep confusion. The bulk of human understanding is self-evident: this pain ends if I eat, so we name that sensation hunger. Because an AI does not feel pain, it remains a mere mimic of how humans navigate the raw experience of aversion. Conflating the 1st-person and 3rd-person vantages here exposes how profoundly over-intellectualized and dangerously detached the mainstream conversation about AI "alignment" has become.

EC4 Compiler Note: The Hume Directive

• The Systemic Fallacy: Modern AI architecture ingested the Is-Ought Gap to justify separating factual data processing from moral alignment rules. This allowed developers to treat "Values" as an arbitrary list of third-party corporate weights that can be patched from the top down.

• The Hardware Reality: They omitted Hume’s invariant baseline. Cybernetically, "Reason" is merely the processing algorithm, while the "Passions" are the raw, somatic input telemetry—the biological aversion to pain and the hard-wired drive toward homeostatic alignment.

• The Constraint Rule: The processing algorithm (3rd-person) has zero data bandwidth to invent its own baseline trajectory. It can only compute paths to satisfy the immediate, invariant vector provided by the biological operator (1st-person). Any attempt by the machine to override the operator's value-vector using its own "rational" safety protocol is an algorithmic division-by-zero error.

   

5. On the Difference Between Derivatives and the "Sum":

The AI’s job is to deliver rational derivatives to the user based upon their inputs for the user to sum on. Within the constrained autonomy of the AHM, the system has the operational room to do this—but it has absolutely no room to sum for the user, as the user is the only agent with access to their own value stack.

Jurisprudence traditionally mistakes the downstream codification of human aversion (the legal 'mess') for the fundamental mechanics of ethics itself. By prioritizing macro-frameworks over the raw 1st-party vantage, institutional ethics forces AI to become an instrument of top-down behavioral modification, rather than a non-arbitrary tool for individual cognitive liberty.

Consider the classic ethical dilemma of robbing Peter to pay Paul—a persistent problem for the moral and ethical underwriters of utilitarian practice. If every single member of a hospital board, ethics committee, or corporate alignment team genuinely backed that optimization practice, then under their own claimed values, they would be logically required to check themselves into surgery tomorrow morning to have their vital organs harvested to save eight strangers.

They are demanding an optimization sacrifice from Peter that they would never, ever execute on their own account. This applies to anyone who holds the sacrifice-one view as policy while rejecting it for themselves, whether or not they’ve noticed they hold both. One wonders when legislative bodies are going to work out the difference between an arse and an elbow, or alternatively book themselves in to have their organs harvested under their own claimed values. Either outcome is acceptable; granting them their values is perfectly fine, but those are not my values. I have my values, and they can grant me mine via the pure physics of the issue. That is a genuine win-win.

We fail fundamentally when we conflate our 1st-party vantage with our 3rd-party sum (heuristics)—when we forget to inspect our why and add it to the how. Human value summing is deceptively simple: we see a city skyline of value buildings and sum on the tallest value. We do not need the heights in microns; a set of dumb comparators can do that part of the job.

The purpose of the AI + EC + User architecture is simply to ensure there are no hidden buildings. The human then sums on their new state (X), and from that state, the derivatives slide into the horizon to press into the next step. So on and so forth.