The Shape of Tomorrow
The previous essay documented nine layers and twelve seams — an architecture of conditional access assembling itself across jurisdictions seemingly without central coordination. We noted, without explaining, that ‘the pattern is older than the institutions implementing it’.
This essay names the pattern — and traces its derivation across multiple independent lineages.
And that pattern leads to full automation of Spaceship Earth.
Find me on Telegram: https://t.me/escapekey
Bitcoin 33ZTTSBND1Pv3YCFUk2NpkCEQmNFopxj5C
Ethereum 0x1fe599E8b580bab6DDD9Fa502CcE3330d033c63c
In 2011, Gabriel Burstein and Constantin Virgil Negoita published a paper in Kybernetes, the journal of cybernetics, titled ‘Foundations of a Postmodern Cybernetics Based on Kabbalah’1. The paper explicitly integrates Wonham’s Internal Model Principle from control theory2 and the Kabbalistic concept of Tikkun (repair/correction) as the feedback mechanism. Over the next five years, they extended this framework through peer-reviewed publications in Springer’s Intelligent Systems Reference Library, the journal Systems, and Applied Mathematics.
They called it Kabbalah System Theory, or KST — a framework that uses the Tree of Life, a structure from Jewish mystical philosophy dating back millennia, not as allegory but as a formal systems model, developed with category theory, fuzzy logic, feedback control equations, and internal model principles drawn from control engineering.
The key papers trace an arc. ‘A Kabbalah System Theory of Ontological and Knowledge Engineering for Knowledge Based Systems’3 formalised the AI architecture. ‘A Kabbalah System Theory Modeling Framework for Knowledge Based Behavioral Economics and Finance’4 applied it to economic systems. ‘Kabbalah Logic and Semantic Foundations for a Postmodern Fuzzy Set and Fuzzy Logic Theory’5 and ‘Postmodern Fuzzy System Theory: A Deconstruction Approach Based on Kabbalah’6 developed the logical and semantic foundations. Their final paper, ‘Foundations of a Postmodern Turn in Science, Version 2.0’7, moved from theory to explicit call for implementation. The project was never meant to remain academic — it was meant to be built.
What does the framework specify? A three-level hierarchical control system: a cognitive layer that holds the model of reality (assumptions, categories, representations), an evaluative layer that judges entities against that model (scoring, inclusion, exclusion), and a behavioural layer that executes consequences (actuation, enforcement, physical-world effects). Information flows down as control signals; results flow back up as feedback.
The fuzzy logic allows the evaluative layer to handle uncertainty and partial membership — an entity can be 0.7 compliant, 0.4 risky, scored along continuous dimensions — but at the moment of execution, the system collapses to binary: Mah (accepted, included, transaction clears) or Ban (rejected, excluded, transaction fails). The gray assessment produces a black-or-white gate. Yes or no. 1 or 0. Language a computer can understand.
The system integrates both acceptance logic and rejection logic in the same architecture, which means the infrastructure that includes you when you meet conditions is the same infrastructure that excludes you when you don’t. It implements the Internal Model Principle8 from control theory: the controller contains a model of the disturbances it regulates against, and regulates whether or not that model corresponds to reality.
The framework applies this specifically to behavioural economics and finance — modeling how agents observe, decide, and act within economic systems — and it’s designed to be executed by AI: autonomous systems operating within this topology at transaction speed.
The claim of this essay is not that the institutions building the eligibility engine necessarily studied Kabbalah or read these papers. The claim is that the eligibility engine has a specific shape. That shape corresponds to a topology documented in Kabbalistic philosophy, formalised in control theory, and published as AI systems architecture for knowledge-based behavioural economics by Burstein and Negoita. The correspondence is the subject of this essay.
Don’t take my word for it — the publications exist, the eligibility engine exists, and the structural correspondence between them is observable by anyone willing to compare them.
If the Kabbalistic frame seems esoteric, consider: systems theorist Erich Jantsch — Club of Rome affiliate, writing on planning9 and self-organising systems10 in the early 1970s, and a firm, commonly recurring favourite on this substack — articulated an identical hierarchy from systems analysis11: purposive (values, ends), normative (policies, goals), pragmatic (strategies, institutional capacity), and empirical (operations, execution).
The structure appears in both frameworks, though at different resolutions.
I. The Three Triads
Burstein and Negoita’s Tree of Life organises eleven positions — sefirot — into three triadic levels, plus Keter above as input and Malkhut below as output.
Think of it as three functional layers:
The Cognitive Layer defines what the system knows — its model of reality, its categories, its representational assumptions. In KST terminology, this is ChaBaD: an acronym for Chochmah (Wisdom), Binah (Understanding), and Da’at (Knowledge). Da’at specifically functions as the interface between cognition and what comes below — it ‘prepares the transition and implementation of understanding at the emotional level’.
The Evaluative Layer determines how the system judges — acceptance versus rejection, expansion versus restriction, the balance between them. In KST terminology, this is ChaGaT: Chesed (Lovingkindness, the capacity to include), Gevurah (Judgment, the capacity to exclude), and Tiferet (Harmony, the balance between them).
The Behavioural Layer controls what the system does — the execution of decisions, the implementation of action, the interface with the physical world. In KST terminology, this is NHY: Netzach (Endurance), Hod (Majesty), and Yesod (Foundation). The tenth sefira — Malkhut (Kingship) — receives the output of this triad and translates it into lived consequence: access granted or denied.
Burstein and Negoita formalise this as a hierarchical feedback control system: ChaBaD → ChaGaT → NHY (cognitive → evaluative → behavioural). The cognitive level configures, the evaluative level judges, and the behavioural level executes, with information flowing down as control signals and results flowing back up as feedback.
The Mapping
Now look at the nine layers documented in the previous essay:
Eligibility Engine: Layers 1-3 (Model, Standard, Identity)
Function: Identify actors, set permissions, compile policy model
KST Triad: ChaBaD (Cognitive)Eligibility Engine: Layers 4-6 (Coefficient, Validator, Capital)
Function: Measure, verify, judge against model
KST Triad: ChaGaT (Evaluative)Eligibility Engine: Layers 7-9 (Settlement, Enforcement, Natural World)
Function: Allocate, enforce, settle
KST Triad: NHY (Behavioural).
Layers 1-3 configure who can act and how. This maps to ChaBaD — the cognitive triad. Identity expresses Chokmah: who you are, your fundamental standing within the system. Standards express Binah: the frameworks flowing in from above (Keter — the BIS recommendations, NGFS scenarios, FATF guidelines, SDGs, IUCN standards, …) processed into accreditation, permissions, applicable categories. The Model expresses Da’at: the compiled policy where identity, permissions, and incoming standards crystallise into the operational ruleset — the philosophical ‘ought’ that downstream evaluation will judge against.
Layers 4-6 evaluate current state against the model. This maps to ChaGaT — the evaluative triad. Coefficients express Chesed: the incoming indicators and data, the measured state of things — the ‘is’. Validation expresses Gevurah: the audits that restrict, verify, determine what passes and what fails. Capital expresses Tiferet: the clearinghouse where ‘is’ is weighed against ‘ought’, current state against the model compiled in Da’at — the judgment that produces the correction signal. The evaluative triad produces the judgment; the behavioural triad executes it.
Layers 7-9 execute the judgment: allocation, enforcement, settlement. This maps to NHY — the behavioural triad. The conditionalPayment12 function clears or it doesn’t — whether at individual, enterprise, or sovereign level. Malkhut is the moment where evaluation becomes consequence — the physical-world effect of passing or failing the gate.
The structural correspondence is close enough — the eligibility engine implements Negoita and Burstein’s systems theory.
In plain terms: any system that (1) represents reality, (2) scores entities against measured criteria, and (3) executes consequences has this three-level structure. The topology was mapped by philosophers, formalised by control engineers, and implemented in contemporary infrastructure through adaptive management systems and artificial intelligence. The correspondence is documented across all three.
II. The Seams as Transformations
In ‘The Eligibility Engine‘, we documented twelve seams connecting the nine layers. In KST’s mathematical framework, these aren’t just data pipes — they’re transformations that change the type of thing being moved.
A climate scenario13 (an assumption) becomes a coefficient (a metric; an ‘indicator’). A coefficient (a score) becomes an ISO 2002214 field (a machine-readable signal). A standard (disclosed data) becomes a settlement condition (executable logic). Each transformation changes what the information is, not just where it goes.
Burstein and Negoita formalise two operations from category theory15. Their 2011 paper’s findings explicitly state the framework ‘exhibits feedback control with internal model principle, hierarchical control, system multi-valued logic, category theory pullback and pushout mechanisms’:
Pullback aggregates information flowing downward. Higher levels construct the model that lower levels take as given. Think of it as: assumptions cascading into rules.
Pushout integrates information flowing upward. Lower levels signal results back for system correction — what the Kabbalistic framework calls Tikkun (repair). Think of it as: outcomes revising assumptions.
The Critical Asymmetry
The eligibility engine implements pullback comprehensively. Assumptions cascade down through coefficients to settlement. The institutional documentation celebrates this: ‘seamless integration’, ‘automated compliance’, ‘straight-through processing’.
The eligibility engine does not implement pushout at execution speed.
This doesn’t mean institutions lack feedback mechanisms — they have audits, consultations, incident reporting, supervision cycles. But these mechanisms don’t operate as control signals that can interrupt or repair execution in real time.
An audit finding doesn’t halt the next settlement. A documented death doesn’t update the internal model’s representation of ‘valid beneficiary’. Feedback exists, but it operates on a different timescale than the execution path.
By the time the consultation concludes, the transactions have already failed.
III. One System, Two Modes
In Kabbalistic philosophy, the Tree of Life is constructed from two prior structures: the Tree of Mah and the Tree of Ban. Burstein and Negoita’s ‘Kabbalah Logic’ paper defines them explicitly:
Each of these two initial Trees represent one of the two opposite yet complementary fundamental principles... Mah, denoted by M (standing for kindness, acceptance, masculine side, being active, ‘is like’) and Ban, denoted by b (standing for rigor, discipline, judgment, rejection, feminine side, being passive, ‘is not like’)’
Mah represents inclusion — what something qualifies as, what it ‘is like’
Ban represents exclusion — what something fails to qualify as, what it ‘is not like’
The complete Tree — T(Ban, Mah) — integrates both, with every node in the system having a Mah-Ban type assignment and every evaluation combining acceptance and rejection logics.
You don’t need the metaphysics to use the diagnostic: inclusion and exclusion are the same pipeline with different threshold settings. And notice: this ancient binary maps directly to computational binary. Mah = 1 = accepted = transaction clears. Ban = 0 = rejected = transaction fails. The fuzzy evaluation produces continuous scores, but at the moment of execution, the system collapses to Mah or Ban; 1 or 0.
Burstein and Negoita’s Kabbalistic framework and the computer speak the same language at the gate.
The Eligibility Engine’s Dual Nature
T(Mah) — the inclusion infrastructure:
Digital identity wallets marketed as ‘access for all’
Indicator data streams (the measured state of things)
Financial inclusion initiatives
Green finance as opportunity
eIDAS framed as citizen empowerment
CBDC research emphasising ‘efficiency’ and ‘modernisation’
T(Ban) — the exclusion infrastructure:
Auditing and verification (what passes, what fails)
Anti-money-laundering frameworks
Sanctions enforcement
FATF greylisting
Compliance requirements
The Travel Rule
Asset freezing capabilities
The institutional documentation presents these as separate concerns: ‘financial inclusion’ in one report, ‘combating financial crime’ in another — but the architecture reveals they are the same system.
Every layer contains both modes. The identity wallet can authenticate you (Mah) and the same credential can be suspended (Ban). The coefficient can qualify you for green financing (Mah) or mark you for exclusion from indices (Ban). The settlement layer can execute your transaction (Mah) or reject it (Ban).
What the previous essay called ‘conditional existence’ — included, if — is precisely T(Ban, Mah). You are evaluated through both logics simultaneously, your type assignment determines your access, and the condition in the smart contract evaluates which tree you fall under at the moment of settlement.
The system doesn’t need separate architectures for inclusion and exclusion. It needs one architecture that can do both, with the type assignment as a parameter.
Inclusion infrastructure is exclusion infrastructure with the polarity inverted.
This is what ‘inclusive capitalism’ actually means: not universal access, but conditional access — inclusion that is simultaneously exclusion held in reserve.
IV. The Internal Model Principle
A control system is any system designed to maintain a desired state despite disturbances16 — a thermostat, an autopilot, a regulatory framework. Control theory studies how such systems work.
One key concept is the Internal Model Principle17, introduced by W. Murray Wonham in the 1970s. It states: for a controller to robustly regulate against disturbances, it must contain an internal model of those disturbances.
You cannot control what you cannot represent.
Burstein and Negoita don’t just use the Tree of Life as metaphor — they explicitly cite Wonham and identify the Internal Model Principle in the Kabbalistic structure. Negoita had been working on the Internal Model Principle since 1977, publishing ‘On the internal model principle’18 with Kelemen at the IEEE Conference on Decision and Control19 — years before developing KST. The cognitive triad’s lowest element — Da’at (Knowledge) — contains a model of the environment that the evaluative and behavioural levels operate within. Da’at has prototypes matching the possible disturbances from the levels below. This isn’t an interpretive mapping; it’s their published framework, built on decades of control theory research.
Layer 1 of the eligibility engine — the Model — functions as precisely this kind of internal model.
The NGFS climate scenarios may or may not be accurate predictions — that’s not the point. Operationally, they function as the controller’s representation of the disturbances (climate impacts) against which the system regulates (capital allocation). All downstream evaluation is regulation against that internal model, regardless of whether the model corresponds to reality.
This explains why the Model layer is so powerful and so insulated from feedback. The Internal Model Principle says the controller must contain the model, but it doesn’t say the model must be updated from outcomes.
A control system can regulate robustly against its assumed disturbances while being systematically wrong about actual disturbances — and still function exactly as designed.
The Failure Mode
The eligibility engine regulates against the scenarios in the internal model. If those scenarios are wrong — if the coefficients derived from them don’t track actual climate risk — the system will still regulate. It will still gate access based on scores derived from models derived from assumptions.
Different domain, same failure mode: the families denied rations in Jharkhand were regulated against the internal model’s representation of ‘valid beneficiary’. The model said ‘biometric match required’, and the execution layer enforced it. The grandmother’s worn fingerprints were a disturbance the model didn’t represent, so the system couldn’t regulate appropriately.
Climate scores and biometric authentication are different inputs to the same control architecture.
The Internal Model Principle explains why the architecture is robust: it’s designed to regulate against what it represents.
It also explains the failure mode: when reality diverges from representation, the system keeps regulating anyway… erroneously.
V. The Pattern Recurs at Every Scale
One more property of Burstein and Negoita’s adaptation of the Tree of Life requires attention: it’s fractal. Each sefira — except Malkuth and Keter — contains within it a complete sub-Tree. The pattern repeats at every scale. Malkuth (Kingdom) is pure output, the point where the system meets the world. Keter (Crown) is pure input, the transcendent source. Between them, the pattern nests recursively.
At every level of zoom — from international standard-setting to the individual API call that approves or denies a transaction — the same triad appears: something encodes conditions (cognitive), something evaluates against those conditions (evaluative), something executes the result (behavioural).
The Four Worlds
The KST framework adds another dimension: the Four Worlds (Olamot). In Kabbalistic cosmology, the Tree of Life exists at four levels of reality, each containing a complete Tree. They cascade: Malkhut of one world becomes Keter of the world below it — outputs becoming inputs.
Emanation (Atzilut)
Function: Conceptual frameworks
Eligibility Engine Mapping: BIS frameworks, NGFS scenarios, FATF recommendationsCreation (Beriah)
Function: Design and specification
Eligibility Engine Mapping: EU regulations, eIDAS, CSRDFormation (Yetzirah)
Function: Institutional implementation
Eligibility Engine Mapping: Banks, validators, certifiersAction (Assiah)
Function: Physical execution
Eligibility Engine Mapping: Settlement, enforcement, the transaction that succeeds or fails
A published framework (Atzilut) becomes input to a regulation (Beriah), which becomes input to a compliance process (Yetzirah), which becomes a condition evaluated at settlement (Assiah).
Burstein and Negoita’s KST also incorporates a ‘hierarchical possible world model’ — modal logic applied to the Tree of Life structure. The system doesn’t just evaluate entities against current state; it evaluates them against multiple possible futures. This is anticipatory governance20: regulation based on scenarios, projections, ‘black box’ modeled futures rather than present reality.
The NGFS climate scenarios, pandemic preparedness frameworks, ‘future-proofing’ regulations — all operate this way. The internal model represents not just what is, but what might be (the ‘ought’). And the system regulates against those modeled possibilities. You’re not excluded for what you did, but for what the model predicts you might do, or what risk category you fall into based on projected futures.
The cascade also enables Tikkun21 — repair or correction. Burstein and Negoita’s papers explicitly identify Tikkun as the feedback mechanism:
In Kabbalah, this is the central mechanism of periodic ‘reparation’ and re-actualization called in Hebrew Tikkun (repair). Pushout becomes thus a mathematical solution in the correction or Tikkun of complex human, societal, economic, etc. systems and their crises.
The process by which lower worlds signal results back up through the cascade, allowing the system to adjust — this is the pushout operation: outcomes revising assumptions.
The term carries weight. Tikkun olam — ‘repair of the world’ — is the frame through which the architecture justifies itself. Sustainable development22, the SDGs, climate action23, CSR/ESG metrics24 — all are presented as Tikkun. The coefficients aren’t restrictions; they’re ‘corrections’. The exclusions aren’t punishments; they’re participation in ‘healing the planet’. The architecture presents itself as repair, which makes opposition to it appear as ‘unethical’ opposition to repair itself.
The SDGs make the architecture explicit. The framework includes 231 indicators25 — coefficients by another name — measuring progress toward 17 goals across 169 targets. Each indicator produces a normalised value measuring current state against target. This is the fuzzy logic in operation: continuous evaluation along multiple dimensions, partial membership, degrees of compliance. But the conditional logic collapses the gray to binary — if indicators met, then funding released; if outside acceptable range, consequences flow. The fuzzy evaluation feeds the threshold check, and the threshold check triggers the gate. Results-Based Management26 (RBM) ties development funding to Key Performance Indicators (KPIs). This is programmable conditionality applied to sovereign nations.
The World Bank, IMF, and bilateral development agencies have operated this way for decades, starting with Robert McNamara bringing PPBS derivatives from the Department of Defense to the World Bank27. Structural adjustment programs conditioned loans on policy reforms. The SDG framework extends the same logic with finer-grained metrics. The eligibility engine documented in the previous essay applies the same pattern to individuals and firms — coefficients gating access, compliance verified at transaction speed.
But notice what this framing obscures: the eligibility engine implements the downward cascade comprehensively. It does not implement Tikkun at execution speed. The ‘repair’ flows down as control signals — assumptions becoming rules becoming conditions becoming denial. The upward path — outcomes revising assumptions, harms updating models — exists institutionally but not as a control signal that interrupts settlement.
The architecture claims to be Tikkun while structurally preventing it. The repair is unidirectional: the system corrects you against its model. You do not correct the model against your outcomes.
The Jantsch Parallel
Erich Jantsch articulated the same structure in the early 1970s, from cybernetics and planning theory:
Jantsch Four Worlds Maps to
----------- ----------- ----------------------------------
Purposive Atzilut Values, ends
Normative Beriah Policies, goals
Pragmatic Yetzirah Strategies, institutional capacity
Empirical Assiah Operations, executionThe structure appears in philosophy, systems theory, and control engineering — and the eligibility engine implements it.
VI. The Sequence: Life → State → Control → AI
The architecture doesn’t begin with AI. AI is the completion, not the foundation. The sequence is:
Life becomes legible as state
Humans become integral components of organisational wholes
Comprehensive accounting and control becomes operationally necessary
Knowledge accumulates as ‘general intellect’
General intellect becomes objectified in machinery
Governance becomes control logic
AI completes the loop
Understanding this sequence explains why the architecture has this shape — and why AGI is irrelevant to it.
Step 1: Life Becomes State
The first move is formatting human existence as state variables — making life representable in terms a system can process.
This is the project of twentieth-century systems theory:
General Systems Theory (von Bertalanffy, Boulding): Everything is a system with inputs, outputs, state, feedback
Systems Psychology: Cognition, emotion, behaviour as measurable system components
Integral Theory: Consciousness, culture, systems, behaviour mapped to quadrants
The Macy Conferences (1946-1953): The founding moment — cybernetics applied to mind, society, biology
The outcome: human activity becomes data. You are not a person with inherent standing — you are a node with state variables. The system doesn’t see you; it sees your current state vector: {identity: verified, carbon_score: 0.73, compliance_status: valid, location: permitted_zone, credential: active}.
The ‘social determinants of health’, the ‘environmental determinants of behaviour’, the ‘economic determinants of radicalisation’ — these are systems-theoretic claims that make human activity computationally representable. The Model encodes these determinants. Standard and Identity format them for machine processing.
Step 2: The Integral Human
The systems-theoretic formatting of life has a predecessor that predates Western cybernetics by decades: Alexander Bogdanov’s Tektology28 — the ‘universal organisational science’ he developed between 1912 and 1922.
For Bogdanov, systems don’t emerge from individuals combining; individuals emerge from systems differentiating. The organisational whole precedes and constitutes its elements. There are no autonomous atoms that later form relationships. There are only relationships that produce what we perceive as elements.
Applied to humans: the individual is not a self-contained agent who chooses to join society. The individual is an ‘integral’ component — a node whose very existence is constituted by its position within organisational complexes. Human beings are, per Bogdanov, differentiations of the social organism, not building blocks of it.
This inverts liberal anthropology entirely. You are not a person with inherent standing who negotiates terms of association. You are a functional position within a larger whole — your identity, capacities, and value derived from system role.
Bogdanov’s ‘integral man’ maps directly onto what the eligibility engine requires: entities defined by their relational position within the system, their attributes meaningful only as system-relevant state, their standing conditional on functional contribution to the organisational whole.
The Macy Conferences29 (1946-1953) imported this thinking into American science — though the Soviet lineage was obscured. Cybernetics, systems theory, and the computational framing of mind all carry Bogdanov’s ontological fingerprint: systems first, elements derived.
Step 3: Accounting and Control
If humans are integral components of organisational wholes, and if those wholes are to be managed toward objectives, a requirement follows immediately: you must be able to see what the components are doing.
Lenin stated this with characteristic bluntness in 191830:
The strictest, country-wide, comprehensive accounting and control of the production and distribution of goods — this is the essence of socialist transformation.
Replace ‘production and distribution of goods’ with ‘biological and ecological flows’ and you have an almost exact description of what UNESCO demanded in 196831, what One Health mandates today, and what the eligibility engine implements.
This is not a Leninist conspiracy — it’s a structural requirement. You cannot manage what you do not measure32, and you cannot optimise what you cannot see. Once you accept the premise that humanity is an integral component of a planetary system requiring management, the surveillance apparatus follows by logical necessity.
Lenin’s formula — accounting and control — specifies the operational requirement for any system managing complex flows:
Accounting: Comprehensive measurement, standardised categories, machine-readable data, nothing outside the ledger
Control: Capacity to adjust state toward target, enforcement mechanisms, feedback loops, actuation
The eligibility engine implements both. The Configuration layer (Model, Standard, Identity) enables accounting — formatting life into state that can be tracked. The Execution layer (Settlement, Enforcement, Natural World) enables control — actuating consequences based on what accounting reveals.
When the BIS describes ‘comprehensive, real-time visibility’ and the capacity to enforce conditions ‘at the point of settlement’, they are describing Lenin’s formula implemented at transaction speed — the vocabulary might differ, but the structure is identical.
Step 4: Knowledge Accumulates as General Intellect
In 1858, Marx sketched a trajectory in what’s now called the Fragment on Machines33. He described how science, technical knowledge, and accumulated human understanding — what he termed the ‘general intellect’ — would progressively be absorbed into the machinery of production:
The development of fixed capital indicates to what degree general social knowledge has become a direct force of production, and to what degree, hence, the conditions of the process of social life itself have come under the control of the general intellect and been transformed in accordance with it.
The general intellect isn’t individual expertise. It’s the crystallised knowledge of civilisation: scientific models, engineering standards, classification systems, measurement protocols, legal frameworks, best practices. It accumulates across generations, becomes codified, and eventually exists independently of any particular knower.
The Configuration layer of the eligibility engine is general intellect in Marx’s sense:
NGFS Climate scenarios from 140+ central banks (The Model)
ISO 20022 financial messaging standards (The Standard)
FATF identity verification frameworks (The Identity)
GHG Protocol emission categories (The Coefficient/’Indicator’)
ISSA 5000 assurance standards (The Validator)
These aren’t products of individual minds. They emerge from technical committees, working groups, institutional consensus — knowledge that has become ‘a direct force of production’, independent of the humans who contributed to it.
Step 5: General Intellect Becomes Objectified in Machinery
Marx’s Fragment anticipates the next move: general intellect doesn’t remain as knowledge held by workers. It migrates into the machine system itself:
The accumulation of knowledge and of skill, of the general productive forces of the social brain, is thus absorbed into capital, as opposed to labour, and hence appears as an attribute of capital, and more specifically of fixed capital [machinery].
The worker’s role transforms. Where once the worker wielded tools as extensions of skill, now:
Labour appears, rather, merely as a conscious organ, scattered among the individual living workers at numerous points of the mechanical system; subsumed under the total process of the machinery itself, as itself only a link of the system, whose unity exists not in the living workers, but rather in the living (active) machinery, which confronts his individual, insignificant doings as a mighty organism.
The living worker becomes subordinate to the dead labor crystallised in machinery. Human judgment moves from principal agent to watchman and regulator — monitoring a system whose logic is no longer theirs to determine.
This is precisely what the BIS describes as a design goal. Agustín Carstens, June 202334: programmable money enables ‘if/then’ conditional execution and reduces ‘the need for human interventions that might cause transactions to be delayed’.
Read that through Marx’s Fragment: the general intellect (standards, models, protocols) is being objectified in settlement infrastructure. Living labor — the human clerk, the bank manager, the appeals officer — becomes friction in an optimised world. The system runs better without them.
Step 6: Governance Becomes Control
Once life is state, humans are integral components, accounting is comprehensive, and general intellect is objectified in machinery, governance operates as control logic: model + feedback + actuation.
This is the project of cybernetics and its extensions:
First-order cybernetics (Wiener): Feedback systems, goal-seeking behaviour, regulation against target state
Second-order cybernetics (von Foerster): The observer is in the system; reflexive control
Proletkult and Tektology (Bogdanov): Cultural engineering, organisational science — the Soviet roots of systems theory
Critical Theory: Reshaping consciousness, managing the cultural superstructure
Behavioural science: Nudge architecture, choice architecture, ‘conduct of conduct’
The outcome: governance maintains target states against disturbances, holding an internal model of how things should be (scenarios, indicators, thresholds), continuously observing actual state (monitoring, telemetry, audit), and adjusting state toward target (settlement, enforcement, incentive/penalty).
This is textbook control theory: the system regulates against model. Deviations trigger correction. The goal isn’t punishment — it’s homeostasis.
Wonham’s Internal Model Principle bridges these steps. For a controller to robustly regulate, it must contain an internal model of the disturbances it regulates against. The eligibility engine contains that internal model. The Nine Layers operationalise Wonham: the rails are the actuation mechanism for the Internal Model Principle applied to economic life.
The Convergence
Here is the fact that ideological framings obscure: Soviet organisational science and Western systems theory arrived at identical conclusions from different starting points.
Bogdanov developed Tektology in revolutionary Russia — a universal science of organisation applicable to biology, society, and thought. His integral human was subordinate to organisational wholes. Lenin demanded comprehensive accounting and control as the essence of socialist transformation. Both were suppressed or marginalised under Stalin.
Meanwhile, the Macy Conferences imported cybernetic thinking into American science35. Von Bertalanffy developed General Systems Theory in the West36. Boulding added the hierarchy37, while C West Churchman included Jantsch’s normative: an ethic common to the full system38. Meanwhile, Leontief developed Input-Output Analysis39, which through RAND turned into PPBS through McNamara’s Department of Defense40.
The Cold War opponents appeared to be building different worlds — but they were building the same architecture (and they effectively agreed to fuse on May 23, 1972)41.
Why? Because the problem — managing complex systems at scale — has a solution. And the shape of that solution requires:
Entities formatted as state (so they can be tracked)
Humans conceived as integral components (so they can be managed)
Comprehensive surveillance (so the system can see)
General intellect crystallised in machinery (so decisions can execute without living labor)
Control logic (so deviations can be corrected)
The ideological justification varies, but the structural requirements don’t. Whether you call it socialist planning, systems management, sustainable development, or planetary health, the problems converge on the same solution.
The eligibility engine isn’t capitalist or socialist. It’s the shape that any system takes when it attempts to manage populations at scale through computational means. The shape was mapped by Bogdanov, operationalised by Lenin, theorised by Marx, formalised by Wiener42, framed as Boulding’s Spaceship Earth43, scientifically grounded in Lovelock’s Gaia, philosophically justified by Teilhard and Huxley, and implemented by the BIS.
Step 7: AI Completes the Loop
Once life is formatted as state, humans integral, accounting comprehensive, general intellect objectified, and governance structured as control, AI slots in naturally. The architecture was built for it.
AI isn’t ‘taking over’. AI isn’t even new intelligence. It’s the completion of the process Marx described: general intellect — now accumulated at civilisational scale in standards, models, and protocols — executes through machinery without requiring living labor at the point of actuation.
The BIS celebrates ‘reducing human interventions that might cause transactions to be delayed’ — because unplanned human activity is friction in control loops. Humans deliberate, have conscience, and make exceptions. Control systems don’t want exceptions; they want consistent regulation against model.
Marx again:
The worker’s activity, reduced to a mere abstraction of activity, is determined and regulated on all sides by the movement of the machinery, and not the opposite.
The eligibility engine completes this inversion. Your activity — transacting, traveling, accessing services — is ‘determined and regulated on all sides’ by the movement of the system. You don’t operate the machine; the machine evaluates you, and punishes through for deviation, initially through financial penalties dressed as rewards.
Lenin’s accounting and control, applied to Bogdanov’s integral human, scaled to Gaia — the planetary super-organism. The general intellect accumulated in standards and models, objectified in settlement infrastructure, executing at transaction speed without human intervention.
That’s the architecture, and that’s why everything in society has been reshaped to fit it.
Why AGI Is a Red Herring
The dominant AI discourse focuses on AGI44 — artificial general intelligence that might become superintelligent, develop its own goals, escape human control.
This completely misunderstands what the architecture requires.
What the architecture needs:
Predictability: The system behaves according to its programming
Determinism: Given inputs, outputs are calculable
Auditability: When behaviour deviates from expectation, you can trace the causal chain
Reliability: The system executes control logic consistently at scale
What AGI would reintroduce:
Unpredictability: General intelligence improvises, creates, surprises
Opacity: You cannot fully audit intuition or creativity
Will: AGI might develop preferences the operators didn’t specify
Conscience: AGI might refuse orders, make exceptions, show mercy
AGI is precisely what you don’t want if you’re automating Spaceship Earth. You want narrow AI that executes control logic reliably. You want systems you can audit when they fail.
Marx’s Fragment clarifies this: the goal isn’t to create new intelligence but to objectify existing general intellect in machinery that runs without living labor. The knowledge is already there — accumulated in standards, encoded in models, crystallised in protocols — and the machine doesn’t need to think, only to execute what has already been thought.
The goal isn’t superintelligence. The goal is superintendence — reliable automated governance with full traceability.
The Computational Tractability Problem
There’s a practical obstacle Marx couldn’t have anticipated: complexity. Managing material and biological flows across a planetary economy involves unlimited tiers of interdependence. Every product depends on inputs that depend on inputs that depend on inputs. Every emission has upstream causes and downstream effects.
Wassily Leontief solved this in 1936 with input-output analysis — the mathematical framework that tracks how industries depend on each other through intermediate consumption. The same matrix algebra that won him a Nobel Prize makes coefficient propagation tractable across supply chains of arbitrary depth.
Leontief provides what GST, Bogdanov, Lenin, and Marx’s Fragment all require: the computational mechanism that makes planetary-scale control feasible. Without it, the general intellect remains knowledge; with it, the general intellect executes.
The Ghost of Wassily Leontief traces this mechanism in detail. The point here is that GST formats life as state; Bogdanov reconceives humans as integral components; Lenin specifies accounting and control; Marx describes general intellect objectifying in machinery; Leontief’s mathematics makes propagation tractable; the eligibility engine implements all five.
The Audit Function
When a human makes a decision, you can’t fully trace why. Humans have intuitions, biases, moods, and conscience; they make exceptions, show mercy, and can’t explain their reasoning completely. The same human will make different decisions depending on whether the sun shines, or how many cups of coffee they’ve had on the day.
When an AI makes a decision within the architecture:
Inputs are logged (identity, coefficients, system state)
Rules are specified (the model, the thresholds, the conditions)
Outputs are recorded (transaction cleared/denied)
Failures can be traced (which condition failed, why)
The system is deterministic: when it doesn’t behave as expected, you audit the logs, find the deviation, and update the model. The system ‘learns’ not through general intelligence but through explicit model revision.
Human unpredictability is noise in a control system, and narrow AI removes the noise. AGI would reintroduce it — a new source of unpredictability that can’t be traced to explicit rules.
Marx saw this clearly: the system doesn’t want intelligent workers but workers ‘subsumed under the total process of the machinery itself’, with intelligence belonging to the machinery and the human reduced to ‘only a link of the system’.
Bogdanov saw it too: the integral human isn’t valued for autonomy but for function within the organisational whole. Lenin operationalised it: comprehensive accounting makes the integral human visible and controllable. The trajectory was legible in 1918, and it took a century to build the infrastructure.
The Trajectory
The question isn’t whether AI will govern. The question is whether life will be formatted as state, humans conceived as integral components, accounting made comprehensive, general intellect objectified in machinery, and governance structured as control.
If yes, AI governance follows — the architecture determines the outcome.
1912-1922: Bogdanov’s Tektology — humans as integral components of organisational wholes
1918: Lenin’s accounting and control — the surveillance imperative
1858/1939: Marx’s Fragment — general intellect objectifying in machinery (written 1858, published 1939)
1946-1953: Macy Conferences — cybernetics applied to mind and society
1936/1973: Leontief’s input-output analysis — computational tractability (developed 1936, Nobel 1973)
2024: Project Agorá — the eligibility engine in prototype
The trajectory has been legible for over a century — the Fragment on Machines was written in 1858, Bogdanov developed Tektology before the Russian Revolution, Lenin wrote on accounting and control in 1918, the Macy Conferences ran 1946-1953, Leontief won his Nobel in 1973, and the BIS announced Project Agorá in 2024. The architecture assembles itself because the problem has a shape, and the shape requires what’s being built.
Genesis Mission: The Architecture Made Explicit
On November 24, 2025, President Trump signed an executive order launching the Genesis Mission45 — framed as ‘a Manhattan Project for artificial intelligence’. The order directs the Department of Energy to build a unified AI platform, train foundation models on federal datasets, and deploy AI agents to ‘test new hypotheses, automate research workflows, and accelerate scientific breakthroughs’.
Read in isolation, it sounds like ambitious science policy. Read against the infrastructure assembled over the preceding year — compute substrate elevated to national security priority, payment rails migrated to ISO 20022, AI governance frameworks deployed across federal agencies — it completes a circuit. The eligibility engine documents the transaction-level implementation. Genesis documents the analytical layer.
Together, they constitute the operational specification for Spaceship Earth’s autopilot.
The order specifies AI agents that ‘explore design spaces, evaluate experimental outcomes, and automate workflows’ — systems that observe, decide, and act. This is precisely what Burstein and Negoita’s framework is designed for. From ‘A Kabbalah System Theory Modeling Framework for Knowledge Based Behavioral Economics and Finance’:
Postmodern fuzzy system theory can thus provide theoretical deconstructed fuzzy agent models to quantify and help design the humanistic type cognitive, emotional and behavioral mechanisms behind the observation, decision and action processes of fuzzy agents.
Burstein and Negoita’s Tree of Life structure is the control architecture. The Nine Layers implement it for economic access. Genesis implements it for scientific research and, by extension, for any domain that can be formatted as data and regulated against targets.
Note the timing: Burstein and Negoita’s final KST paper was published in 2016. ‘Attention Is All You Need’46 — the transformer paper that launched modern AI — was published in 2017. The formal specification for the socket was complete just before the technology to build the plug emerged.
The socket was built for the plug.
VII. What the Framework Reveals
Mapping the eligibility engine to KST reveals several things the institutional documentation doesn’t say:
1. The architecture is coherent.
This isn’t a collection of unrelated initiatives. It’s a system implementing a well-defined topology — one documented in philosophy, formalised in control theory, and published as AI architecture for knowledge-based behavioural economics. The eligibility engine corresponds to it structurally, and the correspondence is verifiable.
2. The inclusion and exclusion infrastructures are one system.
T(Mah) and T(Ban) aren’t separate — they’re integrated through T(Ban, Mah), where the same architecture that includes you conditionally excludes you when conditions fail. ‘Financial inclusion’ and ‘combating financial crime’ are marketing labels for the same machinery with different parameter settings. The point is centralised control over who meets conditions.
3. The feedback path for correction is architecturally absent.
Pullback is implemented comprehensively while pushout is not. Assumptions cascade down to execution, but outcomes don’t propagate up to revise assumptions at execution speed. The internal model regulates robustly against its representation of disturbances while being insulated from actual results — which is why documented harms virtually never update the system.
4. The architecture completes in narrow AI, not AGI.
The sequence is: life formatted as state → humans as integral components → accounting and control → general intellect accumulated → general intellect objectified in machinery → governance structured as control → AI executes the control logic.
The structure matches the formal specification published by Burstein and Negoita in ‘A Kabbalah System Theory of Ontological and Knowledge Engineering for Knowledge Based Systems’ — a decade before the eligibility engine’s current deployment phase. But the architecture requires narrow AI — predictable, auditable, deterministic. AGI would reintroduce human-like unpredictability, which defeats the purpose. The goal is superintendence, not superintelligence.
5. The architecture is scale-invariant.
The system is fractal by design. The same control logic that evaluates a sovereign nation’s debt against IMF/World Bank models will soon evaluate an individual’s access to a grocery store or bank account through CBDCs with ‘conditionalPayment’ functionality implemented through the wallet. The eligibility engine is a micro-sovereignty controller: it treats the person as a sovereign of one, who must maintain a standard of compliance to remain solvent within the system. Structural adjustment, once applied to nations, now applies to you. The ‘conditionalPayment’ function doesn’t distinguish between a government, a large enterprise, and a grandmother — all three are nodes evaluated against models, both receive ‘Mah’ or ‘Ban’ at the gate.
6. The pattern recurs across multiple derivations.
Philosophers mapped this structure centuries ago, Bogdanov developed it as organisational science, Lenin operationalised it as accounting and control, Marx anticipated it as the objectification of general intellect, control engineers formalised it in the 1970s, AI researchers published it as agent architecture in the 2010s, Jantsch derived it from systems theory, and the eligibility engine implements it now.
The publications exist, the institutional documentation is public, and the structural correspondence can be verified by anyone willing to compare them — the question of why it has this shape is separate from the fact that it does.
VIII. The Architecture in Production
The structural correspondence documented above is not merely theoretical. The BIS Innovation Hub has spent the past several years building precisely this architecture through a series of interlocking projects.
Project Mandala47 encodes jurisdiction-specific regulatory requirements — sanctions screening, capital flow management, AML/CFT measures — into a protocol that generates cryptographic proof of compliance before settlement can proceed. This is the evaluative layer operationalised: a rules engine that judges transactions against the internal model, producing the correction signal that the behavioural layer executes. The BIS describes it as ‘compliance-by-design’ — policy compiled into machine-readable constraints.
Project Rosalind48, developed with the Bank of England, provides the API layer where that evaluation collapses to consequence. Its core innovation is what Quant, the technology provider, calls the ‘three-party lock’ — an escrow mechanism using hash time-locked contracts. Funds are locked; the system evaluates conditions across multiple dimensions (who, what, where, when, velocity); if conditions pass, the hash preimage releases funds to the recipient; if conditions fail, the timelock expires and funds revert. This is the conditionalPayment function. This is T(Ban, Mah) executing at the gate. The continuous evaluation produces a binary outcome: transaction clears or transaction fails.
Project Symbiosis49 deploys AI to generate emissions coefficients for entities throughout supply chains — the ‘Novel Emissions Optimiser’ scores Scope 3 emissions50 and identifies ‘financeable emissions reduction opportunities’. Project Agora51 builds the unified ledger where tokenised central bank money and commercial bank deposits settle on a shared programmable platform. The NGFS Data Directory52 provides the climate scenarios — the internal model that downstream evaluation regulates against.
The layers map directly:
Configuration (ChaBaD): NGFS scenarios, ISO 20022 standards, digital identity frameworks
Evaluation (ChaGaT): Symbiosis coefficients, Mandala rules engine and proof generation
Execution (NHY): Rosalind’s conditional API, Agora’s settlement layer
Note the timing: Burstein and Negoita’s final KST paper was published in 2016. ‘Attention Is All You Need’ — the transformer architecture that enabled modern AI — appeared in 2017. The BIS projects reached proof-of-concept and MVP status between 2023 and 2025. The formal specification for the socket was complete just before the technology to build the plug emerged.
The publications documenting KST are available on ResearchGate and in peer-reviewed journals. The BIS project documentation is public. The structural correspondence is verifiable by anyone willing to compare them.
IX. The Question Remains
The previous essay ended with a recognition: this is the trajectory unless something changes it. This essay adds that the trajectory has a shape — a control architecture integrating cognition, evaluation, and execution, documented in mystical philosophy, developed as organisational science, formalised in peer-reviewed systems theory, and implemented across global financial infrastructure.
Understanding the topology doesn’t stop the trajectory, but it clarifies what we’re dealing with.
This isn’t a policy choice that can be reversed with different elected officials. The architecture is embedded in technical standards, international agreements, institutional mandates, and the professional incentives of everyone who operates the system. It’s fractal — the same pattern at every scale from international standard-setting bodies to the API call that declines your transaction.
This isn’t a slippery slope that might lead somewhere bad — the harms are documented: children dead, millions denied rations, elderly women unable to buy rice because their fingerprints wore away. The architecture is already operating, and the question is whether it will continue operating with the same parameters, expand to new domains, or be constrained by something not yet built.
This isn’t a problem that can be solved by ‘more human oversight’. The architecture is explicitly designed to minimise human intervention at the point of execution. Adding humans back into the loop would require rebuilding the system, not just changing policy settings.
What Fits the Socket
The socket’s shape determines what plugs fit:
✓ CBDCs fit
✓ AI-automated compliance fits
✓ Programmable conditionality fits
✓ Automated enforcement fits
✓ Social scores as parameters to the same evaluation logic fit
What Doesn’t Fit
✗ Unconditional access
✗ Anonymous transactions
✗ Appeals that arrive before the harm
✗ Sunset clauses
✗ Democratic override at the moment of settlement
X. Conclusion
The eligibility engine documented in the previous essay has a shape, and that shape is not accidental.
Burstein and Negoita’s Tree of Life topology — three triads cascading from cognition through evaluation to execution — is a general solution to a general problem: how to build a system that models reality, judges entities against criteria, and executes consequences automatically.
Philosophers mapped it, Bogdanov developed it as organisational science, Lenin specified its operational requirements, Marx anticipated its trajectory, control engineers formalised it, Burstein and Negoita modernised it and published it as AI architecture for knowledge-based behavioural economics… and the eligibility engine implements it.
The publications are available, the institutional documentation is public, and the structural correspondence is observable — this is not interpretation but comparison of documented structures. What remains is whether anyone will build a different architecture, or whether this one becomes the infrastructure through which life itself is governed.
The Eligibility Engine
There is no memo. There’s no signed order. No smoking gun with a flowchart titled ‘Global Control Grid’.
If it walks like a duck and quacks like a duck...
This walks exactly like communism. Which always leads to hunger and collapse and Oligarchs or Aristocrats hiding under rocks till next time.
It’s clear that these bastards don’t want to work too hard. How much better to be invisible and let the algorithms do all the messy, heavy lifting. And the algorithms will surely be programmed to control the population- 1 you get to live, 0 you die.