Independent Research Brief · Preliminary Findings
A blinded, pre-registered, calibrated AI-panel study scoring nine models of reality against forty domains of scientific and experiential evidence, and its convergence on consciousness as fundamental.
The Question
Science has gathered an abundance of evidence about the nature of reality, yet what it means remains deadlocked. More than 325 models of consciousness and reality have been proposed.
At the centre of the impasse sits the hard problem of consciousness, arguably science's hardest: why inner experience, the qualia of red, of pain, of joy, exists at all. Rival camps each claim the same data, with no neutral referee between them.
So nine models of reality, chosen to span those families, from physicalist materialism to the simulation hypothesis to consciousness-first idealism, were tested against the same forty fields of evidence, from quantum physics and cosmology to the placebo effect and near-death experiences, to ask which one best explains the whole picture.
The Method
To break the impasse, the evidence was given to an instrument I call the Infoscope: a panel of nine frontier AI systems, each from a different independent provider. Just as the microscope opened up the very small and the telescope the very far, this tool can hold vast, conflicting evidence at once and weigh, by logic, which explanation fits best. It turns a clash of opinion into replicable, quantitative measurement.
The nine were the newest frontier systems available when the study ran, each the current flagship from its provider and spanning the US, Europe, and China: Grok 4.20 (xAI), Gemini 3.1 Pro (Google), GPT-5.4 (OpenAI), Claude Opus 4.8 (Anthropic), Qwen3.6 Plus (Alibaba), Mistral Medium 3.5 (Mistral), DeepSeek V4 Pro (DeepSeek), GLM 5.1 (Z-AI), and MiMo V2.5 Pro (Xiaomi). Using state-of-the-art reasoning, not older models, is deliberate: it is the strongest judgement available, and drawing it from nine rival labs keeps any single house view from steering the result.
Every model of reality was scored against every field by inference to the best explanation. Each was shown only as an anonymised label (blinding), reshuffled on every call, and every judgement was run three times (replication). A separate deductive arm tackled the sharpest question on its own: does the brain generate consciousness, or receive it?
What the Panel Found
The model that ranks first is one in which the physical world is best understood as a virtual experience arising within consciousness itself. Consciousness-first models took the top three places; materialism ranked last on the full evidence; and the result holds across independent cross-checks. The full story is told in the book, and the Compass turns the winning model into a practical tool for life's biggest questions. The chart below lays out how all nine models scored.
Figure. Calibrated fit to the evidence across all forty fields (205 sub-phenomena). 0% reads like a model known to be false; 100% like one known to be true. This is explanatory fit, how well a model explains the data, not the probability that it is true.
It helps to read the chart from the bottom up. Materialism ranks lowest, and the models just above it share its core assumption, that physical matter is what is fundamental: emergentism, panpsychism, the quantum-brain model Orch-OR, and the simulation and Matrix hypotheses. Every one of them stays low. Then comes a sharp jump, scores more than doubling, at the point where a model treats consciousness, rather than matter, as fundamental: dualism, analytic idealism, and, first by a clear margin, VR in Consciousness. The divide is the whole story, and it holds right across the forty fields.
On mainstream science alone (the twelve most conventionally accepted fields), VR in Consciousness still ranks first at 79%, with materialism climbing from last on the full evidence to about 53%. The gap narrows on its home ground, but VR in Consciousness stays clearly ahead.
The Whole-Picture Test
The leaderboard above scores each field on its own, then averages them. In a final stage the panel was shown how each model had scored across all forty fields at once, the per-field results rather than the raw evidence all over again, and asked which model achieves the most with the least: one unifying idea, or a patchwork of special-case fixes.
The consciousness-first model received perfect or near-perfect scores, about three times the average model. Materialism's fit, by contrast, came from accumulated, after-the-fact accommodations rather than genuine prediction. In plain terms, materialism does not naturally explain what we observe: each finding has to be patched in afterward with its own special-case fix, instead of following from the theory itself. That is the signature of a model being stretched to fit, not one that genuinely predicts reality. And the same pattern holds even on the mainstream science alone: VR in Consciousness still comes out on top, and materialism still falls short.
Picture the old parable of the blind men and the elephant. Examine reality one field at a time and you are those men: one holds the trunk and says snake, another the leg and says tree, each right about his own piece and blind to the whole. That is materialism, forever describing the parts. Step back and take in all forty fields together, and a single shape appears that was there all along. The consciousness-first model is the one that finally names the elephant.
What the Winning Model Says About Reality
VR in Consciousness builds on the consciousness-first tradition of analytic idealism, the view that consciousness, not matter, is fundamental. But it draws a far more detailed picture. Where idealism states the general principle, this is the specific shape that emerged once all forty fields of evidence were synthesised into a single account.
In plain terms: consciousness is the only thing that is fundamentally real, and the physical universe, including space and time, is an information-based, holographic, virtual-reality-like experience arising within consciousness. It is made by consciousness, in consciousness, for consciousness to experience, learn, and grow. And you are not separate from it: you are consciousness itself, here having a temporary human experience.
We are not in the world; the world is in us.
The brain, on this view, is a filter and receiver of that experience, not its source. In the language of physics, that one fundamental field is the quantum field; science can measure everything about it except whether it is, in itself, conscious. The model's core argument is that consciousness, the one thing we know from within, is that intrinsic nature.
A preliminary model, offered for scrutiny, not as settled fact.
If what you have just read about our reality and your real self has left you a little uneasy, that is the most natural response in the world, and it passes. You lose nothing, and you remain entirely safe. The world is the same as it was a moment ago; only your view of it has widened, and what comes into focus is kinder, and far larger, than you might fear.
For a fuller, step-by-step picture of what this means for you, click here →
The Brain Question
This is the question the whole debate turns on, and it is where materialism is most exposed. Materialism makes a hard prediction: the brain produces consciousness. So damage the brain and you should damage the mind, quieten the brain and you should dim experience, and where there is no brain there should be no awareness at all.
Yet in thirty-one of the forty fields, where a human experiencer is present, a striking cluster shows the opposite. In near-death experiences, vivid and structured awareness is reported while measurable brain activity is severely reduced or even absent. Near the end of life, terminal lucidity can return sudden clarity to minds that seemed lost to severe brain disease, and dying patients report coherent deathbed visions. New abilities can even appear after brain injury, as in acquired savant syndrome. And under deep psychedelics, experience grows richer just as the brain's main integrative hub, the default-mode network, quietens. The pattern repeats: richer experience, less brain activity, the reverse of what "the brain makes consciousness" predicts. A further, larger group of fields points separately to information and effects that appear to reach beyond the brain altogether.
The remaining nine fields, such as quantum physics, cosmology, and the four-dimensional block universe, have no living brain to measure. There a separate deductive arm weighed the hypotheses by logic alone, and it lands the same way, favouring the brain as a filter and receiver over the brain as a producer by better than two to one: 73.8% versus 26.2% across the nine fields (68.7% versus 31.3% on the seven mainstream fields). Both routes point one way, and it is the opposite of what materialism predicts but exactly what we would expect if consciousness is fundamental: if you are consciousness itself, and the brain is not the maker of your mind but the filter and receiver it works through.
These are preliminary findings, offered for scrutiny, not as settled fact.
Why the Instrument Can Be Trusted
Before scoring any real evidence, the panel was calibrated on test cases with known answers, including a deliberate trap built to fool it (it refused). Calibration also chose the scoring framework that best recovered the known answers.
It also lays to rest a fair worry about any sweeping model, that it explains everything and therefore nothing. Across those test cases, each built around a specific known ontology, VR in Consciousness scored high only where its picture matched the true answer, not everywhere, so its fit is discriminating, not a loose catch-all.
Calibration was only the foundation. Around it the study layered a series of controls, each closing a different route to bias, so the instrument measures what the evidence actually supports rather than what anyone hoped to find:
None of this rests on taking the AI's word for it. Every judgement comes with its full chain of reasoning, the logic and evidence behind each score, written out and archived for anyone to read. The case stands on that reasoning, not on the authority of the machine. An AI's inner workings may be a black box, much as the brain is, but here what matters is fully in the open: you can see exactly how each conclusion was reached, and check it for yourself.
This is AI used as a calibrated measuring instrument, not as an oracle.
By the Numbers
Nine frontier models built by nine independent providers, scoring nine models of reality across forty fields (205 sub-phenomena), validated first on seventeen known-answer fields, every cell blinded and run three times. All told, it ran to more than 18,500 archived AI calls and around 19 million words of recorded reasoning, roughly 240 books' worth.
To my knowledge, one of the largest studies of its kind.
Done the conventional way, by hand, this would be a vast undertaking. Reproducing it would take one researcher an estimated 37 years of focused work, or about 4 to 5 years for the nine independent judges the method requires, at an estimated $3 million in human labour. The AI panel reached the same result in roughly 4 days, for about $600 of compute. These are estimates, yet the gap is not subtle, and it was done with a consistency no multi-year human effort could match, applying the identical rubric and re-randomising the blinding on every one of the 18,570 calls.
Open by Design
Few questions carry higher stakes than what reality actually is, and, bound up with it, what you really are. So being accurate matters more than being right. The full study design, every control, and all the input and output data will be released openly, for independent replication and for others to probe for any flaw. The aim is to converge on the truest model together, on solid data and method, rather than bending the evidence toward a personal, cherished belief about how reality "should" be.
These results are preliminary. Before the book is published, the full study will be repeated once its final fields and sub-phenomena are locked, and re-run on the latest frontier AI models available at that time, for the greatest possible accuracy.
Status: conference-preliminary, under a lean but audited protocol. The nine-panel set is complete and independently verified; a pre-committed fairness refinement and exhaustive post-conference iterations are documented and ahead. Findings labelled preliminary, 12 June 2026. ORCID 0009-0004-5166-6555.
Why It Matters Now
A popular fear runs like this: you are a biological machine, the brain a computer made of meat, and artificial intelligence is the same kind of thing built in silicon. If that were true, consciousness would be nothing but computation, a thing a machine could one day possess and then surpass, leaving you obsolete.
The data points the other way. The simulation hypothesis, that reality is a computer program, scored low, far below the consciousness-first models, and the evidence indicates that consciousness cannot be computed. The reason cuts deep: your true self lies beyond space and time, while a machine, however flawlessly it mimics the outputs of mind, is always contained within space and time. So however far AI advances, it can reproduce what a conscious being produces, even past the point where you could tell the difference, yet never be conscious, just as a map is not the terrain.
At your core you are not a machine but consciousness itself, which lies beyond space and time. No machine can render you obsolete.
The Deepest Implication
An independent, quantitative line of support for a consciousness-first view of reality. Its deepest implication is personal: that you are not a temporary body that switches off, but consciousness itself, having a temporary human experience.
And because consciousness, not matter, is what is fundamentally real, the separation you perceive is illusory: you are interconnected with all that is, and at the deepest level one with it.
Love is the felt sense of that oneness. The completion you keep chasing in outer things is really a longing for that wholeness, a peace found not outside you but beyond the illusion of separation and its drama.
By virtue of what you are, love and peace are not things to attain. They are the very nature of your true, eternal self, beyond space, time, and every possible limitation.
Where This Leads
Finding the most accurate model of reality is only the start. What matters is what it lets us do: understand reality and ourselves more clearly, and put that understanding to use. That is the real mission, not just naming the best map, but learning to navigate by it.
Framed as a familiar VR-like game experience, a reality in which consciousness is fundamental, normally abstract and hard to picture, becomes immediately intuitive. The same simple picture spans the evidence, from quantum physics to near-death experiences, and deepens our comprehension of reality and ourselves, and our ability to teach it to others.
With an accurate picture of reality, you can make sense of life's biggest questions, of experiences like a near-death experience, and of cryptic spiritual ideas, such as non-duality, that are usually too abstract to grasp. The Compass is a free tool in development that helps you decipher what they really mean and tell what is true from what is not.
About the Compass →The Immortal Player (forthcoming, late 2026) tells the full study for the public in clear, accessible language, written as a detective-style investigation that follows the clues across all forty fields and their sub-phenomena.
About the book →And this is only the beginning. A film, an album and an interactive experience are on the way. See what's coming →
One short update a month. You'll be the first to know when the book is released, when the full research data is published, and when the Compass opens.
No spam. No selling your data. Unsubscribe any time with one click.