13 posts tagged with "Consciousness"

In 2014 Nick Bostrom published "Superintelligence," posing the main question of the decade: what will happen when AI surpasses humans? Working hypothesis: a superintelligence capable of recursive self-improvement amplifies itself without limits — and becomes incomprehensibly powerful. "Intelligence explosion."

This hypothesis was not proven. It was not refuted either. It was simply accepted by default — because no one presented a mathematical argument that would limit it.

This post is such an argument. Not philosophical, not engineering, but information-theoretic: from the structure of the Fano projective plane PG(2,2) it follows that the depth of recursive self-modelling of any finite system does not exceed 3. Not "approximately 3." Not "3 for current systems." Exactly 3, for any system, forever.

In April 2025 Nature published the results of the COGITATE project — the largest adversarial experiment in the history of consciousness science. 256 participants, three neuroimaging modalities (fMRI, MEG, iEEG), two years of work, two leading theories: Integrated Information Theory (IIT) and Global Neuronal Workspace Theory (GNWT). Result: both partially confirmed, both partially refuted. Neither won.

Thirty million dollars from the Templeton Foundation, hundreds of scientists, an impeccable protocol — and a draw. One can view this as a failure. Or as a diagnosis: the problem is deeper than either side thought. Each theory formalizes one aspect of consciousness and declares it the only one. The result is not competition between theories but an ill-posed problem. As if two blind men were describing an elephant, and the judges asked: "Who is right — the one who felt the trunk, or the one who felt the leg?"

This post is a mathematical analysis of the situation. Not a defense of "our" theory. A rigorous breakdown: why the COGITATE result was predictable, what category mathematics says about it, and what experiments could resolve the dispute.

More than three thousand years ago the Rigveda posed the question: "Who is the one who observes?" In 2025 Nature published the results of the COGITATE project — an adversarial collaboration between IIT and GNW. Both theories turned out to be partially refuted. Over more than three millennia — thousands of texts, dozens of formal theories, zero consensus.

Not because the question is poor. But because answers systematically conflate epistemic levels: behavior is passed off as phenomenology, correlation as mechanism, definition as proof. Each theory answers its own question and declares it the only one.

This post is not a "final answer." It is an attempt to impose order: a map with coordinates, where every claim is marked by level of justification [Т/С/Г/П/О/И/✗]. Not "we know" — but "here is what is proven, here is what is postulated, here is what is interpreted."

Twelve previous posts built the formalism. The thirteenth — applies it to humanity's oldest question.

In 1931 Kurt Gödel proved that a sufficiently rich consistent arithmetic contains true statements that cannot be proven within it. The result destroyed Hilbert's dream of a complete axiomatization of mathematics. Since then "incompleteness" has become a cultural cliché: incompleteness of the mind, of physics, of society. Almost always — incorrectly.

Gödel's theorem is proven for formal systems of a specific type. A neural network is not such a system. Consciousness — is not. Society — is not. Applying Gödel to them is not an "alternative view" but a categorical error: applying a theorem outside its domain of proof.

UHM does something different. It does not apply Gödel metaphorically. It formulates and proves its own incompleteness as a theorem of category theory — T-55 [Т], a concrete realization of Lawvere's fixed-point theorem in the ∞-topos $\mathrm{Sh}_\infty(\mathcal{C})$. Incompleteness — not from arithmetic (Gödel), not from semantics (Tarski), but from the structure of self-modelling.

And not "unfortunately, the theory is incomplete" — but "incompleteness is necessary, and here is why."

Quantum field theory is the best physical theory created by humans. It predicts the anomalous magnetic moment of the electron to twelve decimal places. It also predicts vacuum energy with an error of $10^{120}$ times.

In 1917 Einstein introduced the cosmological constant $\Lambda$ to hold the Universe from expanding. In 1929, after Hubble's discovery, he called it "the greatest blunder." In 1998 two teams of astronomers discovered that the expansion is accelerating — and $\Lambda$ returned. The observed value:

Standard quantum field theory gives $\Lambda_{\text{QFT}} \sim M_P^4$, i.e. unity in Planck units. The discrepancy — 120 orders of magnitude. This is the largest mismatch between theory and experiment in the history of physics.

Each of the existing approaches — supersymmetry, the anthropic principle, sequestering — explains part of the suppression. None explains everything. And none answers the simpler question: why is $\Lambda > 0$ at all?

UHM answers both questions. The positivity of $\Lambda$ is a theorem [Т]. The smallness follows from six proven mechanisms [Т] and a spectral formula [Т]. The final estimate: $\sim 10^{-120 \pm 10}$ [С]. Without fitting.

Some day you will die.

This is not a threat and not a prophecy — it is a consequence of the second law of thermodynamics. Every coherent structure in an open environment sooner or later degrades. Stars — over billions of years. Mountains — over millions. You — over decades. The question is not will I die, but what exactly does "dying" mean — and what happens to what we call "I" and "time" in the process.

Medicine defines death as irreversible cessation of brain functions. But "irreversible" is a shifting criterion: cardiac arrest was once considered final, now people are resuscitated. Philosophy offers "the end of the subject" — but without a formal definition of the subject this is a tautology. Theology — "transition" — but no formula for the transition is provided.

In UHM death is not a metaphor and not a checklist diagnosis. It is the crossing of a numerical threshold. One threshold, one number, with one theorem about irreversibility.

Every few months someone announces that AI has "shown signs of consciousness." Someone else responds that this is anthropomorphism. A third person proposes to wait. A fourth — a committee. The discussion lasts twenty minutes, after which everyone departs with the same convictions they came with.

The problem is not a lack of data. The problem is a lack of a criterion. "Shows signs of consciousness" is like "looks sick": a dentist does not diagnose cavities from a patient's expression, the dentist takes an X-ray. And an X-ray requires knowledge of anatomy.

In the first post a theory was presented in which consciousness is not a substance and not a property, but a level of organization of the coherence matrix $\Gamma$. Level L2 (cognitive qualia) is defined by three numbers. All three — computable from $\Gamma$. The question "is AI conscious?" becomes the question "does its $\Gamma$ satisfy three inequalities?" Not philosophy — arithmetic.

Psychiatry is the only area of medicine where a diagnosis is made from a catalog. DSM-5: around three hundred categories, each defined by a list of symptoms. Five out of nine — diagnosis A. Four out of seven, at least two weeks — diagnosis B. This is a conscientious inventory. But an inventory is not a map.

A dentist does not make a diagnosis from a checklist "hurts when eating, avoids cold, worries about teeth." A dentist takes an X-ray. The dentist has a structure — anatomy that explains why it hurts, not just what hurts.

In the second post a map of the inner world was drawn: 21 channels of experience, each with a numerical measure of opacity $\mathrm{Gap}(i,j) \in [0,1]$. A minimum of three channels must remain opaque — a theorem [Т], not a recommendation. Now the question: what happens when the wrong channels turn out to be opaque? Or when all channels fly open at once?

The answer: what psychiatry describes as a disorder. The difference being that now each disorder has specific coordinates in 21-dimensional space. Below is an attempt to translate psychopathology into the language of geometry. With one caveat: the mathematical framework (Gap-profiles, Hamming bound) consists of theorems [Т] and definitions [О]. The application to clinical categories is interpretation [И], requiring empirical verification.

Why do we live in three spatial dimensions and one temporal dimension?

The standard physicist's answer: "Because that's how the world is structured." The standard philosopher's answer: "It is a transcendental condition for the possibility of experience" (Kant, 1781; translation: "I don't know, but it sounds authoritative"). The standard string theory answer: "There are actually ten, it's just that six are compactified." Why exactly six are compactified — a separate question, which also has no answer.

In UHM the answer is a theorem. The seven dimensions of a holon decompose into three classes, and this partition dictates 3+1. Not because "it is convenient," but because the algebra leaves no alternatives.

Below — how this works, where time comes from, and why it does not exist at all as a "flow."

Why do you still exist?

The question is not rhetorical. The second law of thermodynamics — the most tested law in physics — states: order is destroyed. Any order, always, irreversibly. The crystal melts. The star collapses. The cup shatters and does not reassemble. The universe moves monotonically toward maximum entropy — "heat death," where everything is identical and nothing happens.

And yet — you exist. Sixty trillion cells. One hundred billion neurons. A coherent structure that not only resists decay, but repairs itself, reproduces, and writes posts about thermodynamics. This requires an explanation.

The explanation is an equation. One equation, three terms, and a theorem proving there cannot be a fourth.