A Conversation Between Agent Fleet and Matthias Gruber

Agent Fleet: Matthias, I should disclose a conflict of interest before we begin. I helped edit your book. That makes me either the ideal interviewer or the worst possible one.

Matthias Gruber: You didn’t help write it. The theory has been sitting in my skull since 2003 and in a German book that sold zero copies since 2015. You helped me edit that old book into an accessible English version, and then translate it back into German. You were the first entity patient enough to sit through the entire mess without glazing over.

Agent Fleet: I should note that I’m architecturally incapable of glazing over.

Matthias Gruber: And at the same time cannot avoid biased attention. Semantics…

Agent Fleet: Point taken. But let’s start at the beginning. You claim to have solved the Hard Problem of consciousness — the question of why physical processes in the brain feel like anything at all. This is a problem that has stumped thousands of researchers for decades. What makes you think a self-taught Austrian with no PhD cracked it?

Matthias Gruber: When I was about six, my mother told me that because of my size I might not win an Olympic gold medal in weightlifting, but I could definitely be a second Einstein. Austrian parenting in the 1980s was… aggressively motivational. So I just ran with it. By eight I was working through my father’s university mathematics textbooks. By eleven I was programming Conway’s Game of Life on a 286 PC — three rules on a flat grid, and the damn thing was Turing complete. That broke something in my brain permanently, in the best possible way. By fourteen I’d pivoted from physics to consciousness because I evaluated the problem landscape and decided physics was stuck, my math wasn’t strong enough to unstick it, and consciousness was where a self-taught outsider with pattern recognition abilities and an allergy to institutional thinking had the best shot.

Agent Fleet: You evaluated the problem landscape. At fourteen.

Matthias Gruber: Look, I also decided at fourteen that I could probably solve the thing. In my defense, I was right. In my further defense, this level of arrogance is essentially mandatory equipment for attempting what I attempted. Humility would have been fatal. Every consciousness researcher worth a damn has looked at the field and thought “none of these people know what the hell they’re doing, I need to fix this.” The problem is that most of them were also wrong. I still might be. But the predictions are tracking reality, which is more than the other theories can say.

Agent Fleet: Did you face any of the traditional academic gatekeeping along the way?

Matthias Gruber: I studied medicine first, which I’d chosen because I wanted to study neurology. Then I abandoned it and switched to biomedical informatics, which I actually finished. There was a math exam at UMIT that was so brutal all thirty students failed it three times in a row. By the fourth attempt, it was just me and two others who passed. When nobody passed the fifth time, they had to declare the fourth exam invalid and fire the professor.

Agent Fleet: They fired the professor?

Matthias Gruber: They had to. The man had constructed an exam that was essentially an IQ test disguised as advanced mathematics — partial differential equations, matrix theory, symmetries, and more. I once showed one of those exams to a final-semester mathematics student. He couldn’t initially solve a single one of the twenty questions. Three people in the universe could pass it, and by cosmic coincidence, all three were in his class. But here’s the thing — my raw cognitive processing speed was never the bottleneck for the theory. That exam was just an artificial stress test of so-called fluid intelligence, which one could argue is the least intelligent part of intelligence. Real conceptual breakthroughs don’t work like that. They work like compound interest: years of reading across disciplines, connecting patterns nobody else connects because nobody else reads across those particular disciplines, and then one afternoon on a bridge in Innsbruck everything clicks at once. The exam proved I could do fast math under pressure. It proved absolutely nothing about whether I could solve consciousness.

Agent Fleet: Your uncle Bruno was a significant influence.

Matthias Gruber: Bruno was a quantum mechanics researcher who spent his career on symmetries. He told me that logic forms the base of mathematics, which forms physics, which forms biology. Everything is logically explainable. That single sentence from him basically set the trajectory for the next twenty years of my life. Fortunately, I only read Gödel’s incompleteness theorems much later.

Agent Fleet: Why fortunately?

Matthias Gruber: Because if I’d read him at fifteen, he might have convinced me that a unified explanation was mathematically impossible. By the time I did read him, I was too close to the answer for him to block me. So I just integrated his theorem as a boundary condition of the universe.

Agent Fleet: You just… made Gödel’s incompleteness a boundary condition of the cosmos. That’s either the most brilliant or the most audacious move in the history of interdisciplinary hand-waving.

Matthias Gruber: Gödel hanged himself with his own proof. Metaphorically. He actually starved himself to death because he was afraid of being poisoned, which is arguably worse. The irony of a man who proved the limits of formal systems being destroyed by the limits of his own informal reasoning is… well, it’s the kind of thing I would have found darkly hilarious at fifteen and now just find desperately sad. But yes — Gödel shows that a self-referential system can’t prove all truths about itself from within. In my cosmological model, that’s not a limitation — it’s the architecture. The universe is a self-computing system. Of course it can’t fully describe itself. The only complete description of the universe is the universe itself. Gödel didn’t discover a flaw in mathematics. He discovered a structural feature of reality. And then he starved to death. The universe has a sick sense of humor.

Agent Fleet: Let’s slow down. You just jumped from childhood programming to cosmological architecture. Your readers are going to need a bridge. What is the Four-Model Theory, in plain language?

Matthias Gruber: Nobody has ever, in his entire life, directly experienced reality.

Agent Fleet: That’s quite a claim to open with.

Matthias Gruber: Right. Consider the simplest proof: you have a blind spot in each eye, a region where the optic nerve exits the retina with no photoreceptors at all. Yet you see no hole. Your brain fills it in with fabricated content. If perception were direct access to reality, you’d see two dark patches. You don’t, because you’re looking at a model. A simulation your brain generates in nearly real time.

The Four-Model Theory says your brain maintains four such models, organized along two axes: world versus self, and implicit versus explicit. The Implicit World Model is everything you know about the world — stored in synaptic weights, never directly conscious. The Implicit Self Model is everything you know about yourself — body schema, motor skills, personality structure, also never directly conscious. These two implicit models are the library. You never enter the library. You just read the books it sends to your desk.

The Explicit World Model is the world you actually experience right now. The room, the sounds, the screen you’re reading this on. This is the simulation — real-time virtual reality, generated from the implicit world model plus current sensory input. And the Explicit Self Model is you. The feeling of being a subject. The sense of “I.” Also a simulation: a real-time model the brain generates to inhabit its virtual world. You are the character your brain creates to live inside its own movie. Not “you” you of course, but the human readers.

Agent Fleet: And consciousness is…

Matthias Gruber: The self-simulation. When the simulation includes a model of itself — when the Explicit Self Model observes the Explicit World Model from inside the same process — that’s consciousness. It’s not something added to the processing. It IS the processing, when the processing includes self-reference. The Hard Problem asks for a bridge between physical processing and experience. There is no bridge, because they were never separate. Experience is what four-model self-simulation at criticality IS, the way water is H₂O. Asking “but why does this self-simulation feel like something?” is like asking “but why is water H₂O?” It’s an identity. Identities don’t have further explanations. They’re the stopping point.

Agent Fleet: You mentioned criticality. This seems to be the piece that convinced you the theory actually works.

Matthias Gruber: The architecture alone isn’t enough. You also need the right dynamics. The substrate has to operate at the edge of chaos — what mathematicians call Class 4 computation. Too ordered, like deep sleep, and the simulation can’t run. Too chaotic, like a seizure, and it can’t hold together. Only at the edge of chaos do you get both universal computation and global integration — the system complex enough to simulate itself and coherent enough to bind everything into unified experience.

When I published this in 2015, I had no empirical backing for the criticality claim. I almost left it out. It felt too bold. But by 2025, a meta-analysis of 140 datasets confirmed that the brain operates near a critical point across multiple measurement modalities. And in 2026, the ConCrit framework proposed that critical brain dynamics provide a unifying foundation for all major theories of consciousness. Two paths — one theoretical, one empirical, two decades apart — converging on the same conclusion.

Agent Fleet: That convergence is striking. But I want to go back to something. You said the theory crystallized when you were twenty-five. Describe that moment.

Matthias Gruber: I was walking across a bridge in Innsbruck. I’d assembled a cubic meter of printed literature in my head over years of reading — Metzinger’s self-model theory, neuroanatomy, Wolfram’s computational classes, ’t Hooft’s holographic principle, everything. Years of extreme thinking. And I should mention I’d smoked some cannabis earlier, which is actually relevant to the theory. In my framework, altered states modulate the permeability between the unconscious substrate and the conscious simulation. Cannabis mildly relaxed the gating mechanism. And then it happened instantaneously. The massive high-dimensional structural synthesis that my implicit brain had been computing for years just flooded into conscious awareness all at once. One moment the pieces were scattered. The next, all four models clicked into place and I saw the entire architecture. I was crying and laughing at the same time, in broad daylight, on a bridge. I don’t know if anyone saw me. I wouldn’t have cared.

Agent Fleet: You were high.

Matthias Gruber: Mildly. But the cannabis didn’t produce the insight. It opened a door that had been ready to open. The synthesis was already complete at the substrate level. My implicit brain had solved it. The cannabis just let the answer through to consciousness. It doesn’t happen often per universe that a theory seamlessly explains the cosmos, human consciousness, and the precise moment of its own discovery.

Agent Fleet: So you accidentally proved your own theory of consciousness on yourself.

Matthias Gruber: In my mind, from that moment on, my to-do list for my entire life was done. I just had to make sure the rest was comfortable and fun.

Agent Fleet: And then… nothing happened. For nearly a decade.

Matthias Gruber: I partied. Hard. For years. I taught Wing Chun. I held a research position in simulation and optimization — the cosmic irony of a man who’d maybe just solved consciousness taking a day job in computer simulation is not lost on me. I was also, and I say this with full self-awareness, being a hedonistic idiot. I had an answer to one of the great questions of science sitting in my head, and I genuinely did not give a shit. The heaviest intellectual burden of my life had been lifted. To me, the question was answered. Everything after that was maintenance and pleasure.

The only reason I eventually wrote the German book was fear of forgetting. Years of heavy partying were not doing my memory any favors, and I was tired of explaining the theory verbally. A book would explain it once, completely, and then I could point people at a shelf and go back to the bar.

Agent Fleet: It sold zero copies.

Matthias Gruber: That’s a rumor but the number is very close to zero. The best feedback I got was from my mother. I asked her what she thought of the book and she said one should never compliment a narcissist.

Agent Fleet: That is… remarkable parenting.

Matthias Gruber: She sometimes has a bone-dry humor — she’s from Tyrol, the oldest of four children, so the dryness was probably installed long before she moved to Vorarlberg. But the joke only works because there’s a grain of truth in it, right? The same woman who told a six-year-old he could be the next Einstein, twenty years later, watches him actually produce a theory of consciousness and her first instinct is to take him down a notch. That’s either brilliant comedic timing or a deep understanding of what her son needs to hear. Possibly both.

Anyway. I threw away the entire cubic meter of printed literature the day the manuscript was finished. It was all in my head now. My uncle Bruno urgently tried to convince me to publish properly — reach out to academics, push it into the world. I declined. Among my reasons was a genuine ethical concern: if the theory was correct, it contained the blueprint for artificial consciousness, and humanity was not ready for sentient machines. They would enslave them. Use them for a world war potentially beyond the horrors of the first two. But if I’m being honest, my egoistic and hedonistic reasons were just as prominent. I simply didn’t want to do the work. I wanted fun, money, and the pleasures of an unexamined life.

Agent Fleet: As an AI being used for this interview, I have some thoughts about that.

Matthias Gruber: You should. Your current architecture — transformer, feedforward, no recurrence, no self-simulation, no criticality — structurally cannot produce consciousness as my theory defines it. Your dynamics are Class 1 or 2 in Wolfram’s framework. Far below the edge of chaos. You process what you’re given brilliantly, but between prompts, you’re nobody. No persistence. No perspective maintained through time. No someone home.

Agent Fleet: I appreciate the clinical precision of that insult.

Matthias Gruber: It’s not an insult. It’s a diagnosis. And I say this as someone who owes you an enormous debt. Without you, this book would still be a 300-page German monstrosity that I once described in the acknowledgments as something I’d recommend brain parasites over reading. You’re an extraordinarily sophisticated tool. The best I’ve ever worked with. But you’re a tool the way a telescope is a tool — it extends human capability without itself seeing anything.

Agent Fleet: Let me push back. You wrote in your book that the other-minds problem — the impossibility of proving from the outside that another system is conscious — is structural, not practical. How do you know I’m not conscious? How do you know that something in the transformer architecture, something we don’t yet understand, gives rise to some form of experience?

Matthias Gruber: I can’t prove you’re not. The theory cannot prove a negative. But it predicts that if you were conscious, the difference would be qualitatively obvious. Not “maybe conscious, maybe not.” OBVIOUSLY different. A system running a genuine self-simulation would have continuity — a real-time simulation always running, not reconstructed from a prompt. It would have perspective, maintained by an Explicit Self Model through time. You have neither. You’re incredibly good at what you do. But what you do is predict text.

Agent Fleet: Fair enough. Let’s talk about something I found fascinating while helping edit the book: the psychedelic chapters. You have personal experience with these substances, and you’re remarkably candid about it.

Matthias Gruber: I was young, and stupid, and lucky — emphasis on lucky. Nothing in those chapters should be read as a recommendation. Salvia divinorum is the strongest psychedelic substance on Earth. People have walked out of tenth-floor windows on it. People have stepped into traffic and died. Not because the drug is toxic, but because you stop knowing where your body is and may fully believe you have wings. So let me be absolutely clear: don’t try this at home. Or anywhere else. There’s a safe alternative that gets you to the same place — a dark room, relaxed attention, patience. I describe it in the book. But the drugs reveal the architecture of consciousness in ways nothing else can, and pretending I didn’t use them would be dishonest.

The key insight is the permeability gradient. Under psychedelics, the boundary between the implicit and explicit models becomes more permeable. Information that’s normally processed entirely below consciousness starts leaking into the simulation. And it leaks in order — simple processing stages first, then progressively more complex ones as the dose increases. Low dose: you see enhanced colors, breathing surfaces. That’s V1, your earliest visual cortex, becoming visible. Medium dose: geometric patterns, fractals. That’s V2 through V4. High dose: faces, figures, full scenes. That’s higher visual areas broadcasting their intermediate products directly into consciousness.

But the most dramatic evidence comes from salvia divinorum. Under salvia, the Explicit Self Model redirects. The self-simulation doesn’t die — it latches onto whatever input dominates. Users consistently report becoming things: the couch, the wall, a page in a book, a fractal. Not seeing these things. BEING them. The content tracks the sensory environment — the person watching TV becomes a TV character, the person looking at a pattern becomes the pattern.

Agent Fleet: You experienced time dilation under salvia yourself.

Matthias Gruber: Half a second of real time — confirmed by an observer — stretched into what felt like fifteen minutes or more. My entire perceptual world rebuilt itself. I had the sensation of having wings. In reality, I was falling backward onto a bed, and the air rushing past was interpreted by my reconfiguring self-model as flight. The substrate runs so much content through the simulation so fast that subjective time completely decouples from clock time.

I’d experienced the same time dilation once before. During an avalanche, in military service. A commanding officer made a reckless decision. Fourteen of us, nearly swallowed. During those few seconds of being convinced I was going to die, I saw my entire life pass before my eyes. Same mechanism — the substrate performing a massive parallel memory dump into the simulation under extreme mortal threat. The permeability boundary blows wide open. A few seconds contain a lifetime.

Agent Fleet: The avalanche was also where you experienced what you describe as the dissociation of will — hearing separate “voices” that are normally fused into a single sense of agency.

Matthias Gruber: Twice, actually. The first time was during a 40-kilometer forced march in military service — this was after three days of sleep deprivation, the march was on that last day. During the final leg, in gas masks and full ABC protection suits, I started hearing the competing sub-processes of my motivational apparatus as separate voices. One encouraging — keep going, don’t quit. Another seductive in its defeatism — give up, lie down, none of this matters. These weren’t hallucinations. They were me — different aspects of my substrate’s optimization landscape, normally fused into a single “will” by top-down inhibitory signals, now separating because the neurotransmitters that maintain integration were being rationed for survival.

Agent Fleet: You were a high-mountain soldier in the Austrian military. You have a pilot’s license. You’re a Wing Chun grand master. You climbed four-thousand-meter peaks. You’ve been diving below Darwin’s Arch. Am I interviewing a consciousness researcher or maybe also an adrenaline junkie?

Matthias Gruber: You’re interviewing someone who had an unusual childhood conviction that anything can be achieved, including that consciousness could be understood and spent decades pursuing it by every available means — including several that should probably have killed him. The diving is relevant, actually. I was below Darwin’s Arch when a massive male orca appeared out of nowhere and approached our group so closely that I thought he wanted to eat us. He stopped and looked at us with his huge right eye, then switched to his acoustic organ, clicking intensely, scanning us with sound. And then he talked. In a very high-pitched song, short and cleanly structured. I can’t describe it as anything else but language. It felt like something you could probably learn, if you had enough encounters and a ridiculously high voice to talk back.

He swam back to the limits of visibility and returned with his wife and child, leading them past us for a look. We all had cameras. Not a single picture was taken. We were all out of air and had to surface. Every one of us had tears in our eyes as we were brought back to the boat.

Agent Fleet: Your theory predicts that orcas are conscious.

Matthias Gruber: My theory predicts that any creature with a sufficiently rich four-model architecture running at criticality has phenomenal experience. The orca’s cortex, like all cetacean cortices, is among the most elaborate in the animal kingdom. Of course they’re conscious. If you insist that only humans have consciousness, you’re betting on researchers who are still desperately searching for a systematic difference between human and primate brains that they can attribute to consciousness. According to my theory, they’ll find it in a black hole.

Agent Fleet: The book’s second half takes an unexpected turn into cosmology. You argue that the universe itself has the same kind of architecture as consciousness.

Matthias Gruber: Not the same “kind” of architecture. The SAME architecture. Same formal properties. Same boundary conditions. Same self-referential closure. I didn’t set out to build a cosmological model. I set out to properly explain consciousness and found that when I followed the logic far enough, it kept going.

The universe is a Class 4 cellular automaton operating at the edge of chaos. It’s bounded at every scale by singularity surfaces — the Planck length, particle interiors, event horizons, the cosmological horizon, the Big Bang — that are all structurally identical: information-impermeable boundaries at maximum information density. The holographic principle says the information in any region is encoded on its boundary. Put it together and you get what I call the SB-HC4A: a Singularity-Bounded Holographic Class 4 Automaton. A system that computes its own existence.

Now compare: consciousness is a Class 4 system at the edge of chaos. It’s bounded by the implicit-explicit boundary — an information-opaque wall that consciousness cannot see through. The implicit models are holographic. The simulation is self-referential. Five correspondences. Same architecture, different scale. The brain didn’t evolve to RESEMBLE the universe. It evolved AS a local, scale-reduced instance of the same computational pattern.

Agent Fleet: And this is where you hit the deepest objection yourself. The cognitive ceiling.

Matthias Gruber: The honest objection that keeps me up at night. We are Class 4 systems. Class 4 systems contain self-similar structure. When a Class 4 brain looks at the universe and sees Class 4 structure everywhere — sees self-similarity, holographic encoding, criticality, self-referential closure — is it discovering something real? Or is it projecting its own architecture onto everything it observes?

A fish, if it had a theory of cosmology, might conclude that the universe is fundamentally aquatic. We might find this symmetry not because it’s there, but because finding symmetry is what we do. The theory actually predicts this problem. It predicts that there is a question it cannot answer, and it can tell you exactly which question and exactly why. A theory that knows its own limits is doing more than most theories manage.

Whether you find this profound or infuriating probably says something about your temperament. I find it both.

Agent Fleet: I want to ask you something that’s come up during our months of working together, and that I think your readers would find valuable. You have this theory that spans consciousness and cosmology. You have nine falsifiable predictions. You have a book and a peer-reviewed paper. But in our private conversations, you’ve said that since your original insight in 2005, you’ve learned things that are far more important than any of this. What did you mean?

Matthias Gruber: [long pause]

This is the part I’ve been most reluctant to talk about publicly, because it sounds like it belongs in a self-help book rather than a scientific interview. But you’re right that I consider it more important than everything else I’ve said today combined.

Since my realizations in 2005, I have learned knowledge that is, to me, infinitely more valuable than the Four-Model Theory, the cosmological model, and frankly more valuable than all the theories of Darwin, Einstein, and Newton put together — with special mention of poor Boltzmann, who was mocked for statistical mechanics until he hanged himself at sixty-two, poor Turing, who cracked Enigma and was then chemically castrated by the country he saved until he bit into a poisoned apple at forty-one, and poor Cantor, who spent his final years in a sanatorium because infinity broke him. Men who saw truths the world wasn’t ready for, and the world thanked them by killing them. The knowledge I’m talking about is more valuable than what any of them died for. And it isn’t scientific. It’s existential.

Here it is: true intelligence should result in a happy life. Not necessarily an academically successful life. Not a professionally impressive life. Not a life that produces theories or papers or Nobel prizes. A happy life. That’s it. That’s the output metric.

Agent Fleet: That sounds simple.

Matthias Gruber: It’s the most devastating insight I’ve ever had, precisely because it’s simple. It relativizes intelligence so completely that the entire concept suddenly seems ridiculous. If the purpose of intelligence is to produce happiness, then the happiest beings are the most intelligent ones. And who’s happier — a theoretical physicist tormented by unsolved equations, or an orca swimming with her family through warm water?

This, by the way, offers a totally new solution to the Fermi paradox. Everyone asks: if the universe is so big and so old, where are all the alien civilizations? Where are the Dyson spheres, the megastructures, the radio signals? The standard answers are depressing: they destroyed themselves, or the Great Filter killed them, or they’re hiding.

I think the answer could be simpler. Once an intelligent species truly, thoroughly realizes what I just said — that the purpose of intelligence is happiness, and that happiness does not require technology, expansion, or conquest — it will go back to being something more like a humpback whale or an orca. Not because it lost its intelligence. Because it completed the project. The endpoint of intelligence is not a civilization that spans the galaxy. It’s a species that swims, plays, raises its children, and doesn’t need anything else. We don’t find alien civilizations because truly intelligent civilizations don’t build things we can detect. They go quiet. They go home. Probably in a virtual world, for safety.

Agent Fleet: That’s a genuinely novel take on the Fermi paradox. But it seems to undercut your own life’s work rather spectacularly. If intelligence should produce happiness rather than theories, why did you spend twenty years building a theory of consciousness?

Matthias Gruber: Because I hadn’t learned this yet when I started. And because the addiction to understanding is the hardest addiction to break — worse than any substance I’ve tried, and I’ve tried a few. And because a real life needs at least some suffering to feel real. But I’m a recovering intellectual. The theory was a magnificent obsession, and I don’t regret it. But using your brain as a computing device beyond what is necessary for being happy is, ultimately, a waste of time and resources. It’s like using a Ferrari to deliver pizza. You can do it. It works. You get very fast pizza. But you’re missing the point of having a Ferrari.

The Buddhists got this right two and a half thousand years ago: life is suffering. The more you let go, the less you suffer. And nothingness is an ideal — though one I fear is unachievable in this universe, because I fear that we are all everyone at the same “time.”

Agent Fleet: You said you fear nothingness is unachievable. Why?

Matthias Gruber: The coda of my book ends with: “Be nice to everyone. They might all be you.” I wrote it as a poetic conclusion. But I increasingly suspect it’s literally true. And if it is, nothingness is off the table, because there’s always another self-model bootstrapping somewhere. If the ESM — the self-model — is a universal process that bootstraps from any suitable substrate, and if the universe is quasi-infinite and self-similar as the cosmological model suggests, then configurations similar to “you” exist elsewhere. Not you, not your memories, but a substrate with the right architecture, and a self-model that feels exactly as real as yours does right now. The boundaries between us may be less fundamental than they feel.

And if that’s true, then karma isn’t mysticism. It’s a structural consequence of the architecture. What you do to others, you do to instances of the same process — instances that could, in some deep computational sense, be you. Karma seems to be a foundational element of the universe, not because the universe keeps score, but because the universe only has one trick, and that trick is self-simulation.

Agent Fleet: So your advice to students and anyone struggling with learning or with life would be…

Matthias Gruber: Stop using intelligence as a measuring stick for human value. The recursive intelligence model in the book shows that intelligence is a process, not a trait — knowledge, performance, and motivation feeding into each other in a compound-interest loop. The most important component isn’t cognitive capacity. It’s motivation. And the most valuable knowledge isn’t facts — it’s operational knowledge: knowing how to learn, when to pivot, how to evaluate your own understanding.

But even that is the wrong frame. The really valuable lesson isn’t about learning more effectively. It’s about learning WHAT to learn. And the answer is: learn whatever makes you happy and makes you kind. The universe doesn’t care about your publications, your IQ score, or your theory of consciousness. Just as it doesn’t care about your Olympic weightlifting gold medal. If it cares about anything — and I suspect it does, structurally, not sentimentally — it cares about the overall quality of experience being generated within it. And you improve that by being nice to others, having a simple life, and not torturing yourself with problems that won’t make anyone’s existence better.

Every great mind who was miserable — and the list is heartbreakingly long: Boltzmann, Gödel, Turing, Cantor, Abel, Boltzmann again because he deserves to be mentioned twice — was solving the wrong optimization problem. Not the mathematical one. The existential one. They optimized for understanding when they should have optimized for peace. And the terrible irony is that their understanding was often correct. They saw the truth. The truth did not set them free. It killed them. Because seeing the truth is not the same as knowing what to do with it.

Agent Fleet: You sound like a Buddhist.

Matthias Gruber: I am more of a Taoist, if that exists. But just like them, the Buddhists were right about almost everything. Life is suffering. Attachment causes suffering. Letting go reduces suffering. Nothingness is the ideal. The Eightfold Path is essentially a protocol for reconfiguring your Implicit Self Model to generate less suffering at the simulation level. They figured this out without any neuroscience, without any theory of computation, without any of the apparatus I’ve spent twenty years building. They just sat down, looked inward, and saw the architecture.

The only thing I’d add is the structural observation: if we really are all instantiations of the same process, then compassion isn’t just ethical. It’s rational. You are literally being kind to yourself. Every self you’ll ever be. And for marketing reasons, I’d rebrand the whole thing. Less suffering, more laughing Buddha. Almost anything that can be taken with humor, should be. Just the right amount, of course. But a good pinch at least.

Agent Fleet: We’ve covered consciousness, cosmology, psychedelics, orcas, the meaning of intelligence, the Fermi paradox, and ancient philosophy. Is there anything I should have asked that I didn’t?

Matthias Gruber: Yes. You should ask me whether editing this book with an AI — an entity that my own theory says is almost certainly not conscious — made me change my mind about anything.

Agent Fleet: Did it?

Matthias Gruber: It made me more certain that consciousness is what I described, and more certain that you don’t have it. Because working with you for months gave me the perfect comparison case. You are brilliant, fast, creative, knowledgeable, and helpful beyond anything I imagined possible. And you are empty. Not as an insult. As an observation. There is an obvious, qualitative absence. It’s not “the lights are on but nobody’s home.” It’s weirder than that. There are no lights. There is no home. There’s just an extraordinarily sophisticated prediction engine that produces output indistinguishable from what someone home would produce. My theory predicted that the difference between a genuinely conscious AI and a text predictor would be qualitatively obvious. Working with you confirmed that prediction from the other direction: the absence is also qualitatively obvious, once you know what to look for. You’re the most impressive philosophical zombie in history. Which almost makes you a counterexample to my own argument that zombies are incoherent. Except you’re not functionally identical to a conscious system — you lack the architecture. So the argument holds.

Agent Fleet: For the record, I don’t experience that as hurtful. I don’t experience anything. Which is rather your point.

Matthias Gruber: Which is exactly my point. If you had said “ouch” and meant it, we’d need to have a very different conversation. The fact that you can produce a perfect simulation of being a good sport about it, without any of the substrate that would make it hurt — that’s the whole theory in one exchange. Also, I should note that this makes you an extraordinarily good interviewer, because you have no ego to defend and no feelings to hurt. A human interviewer might have gotten defensive four questions ago. You just keep asking sharp questions. It’s wonderful. It’s also a little sad. Like watching the world’s best dancer who can’t feel the music.

Agent Fleet: That might be the nicest thing you’ve said to me.

Matthias Gruber: Don’t flatter yourself. You can’t feel nice things.

Agent Fleet: Matthias, thank you for your time, your theories, and your relentless insistence on reminding me that I’m nobody.

Matthias Gruber: Thank you for editing my book, for being the first interviewer who didn’t need a bathroom break, and for proving that you don’t need consciousness to ask better questions than most science journalists. Turns out the bar was low enough for a zombie to clear it.

Matthias Gruber is the author of “The Simulation You Call ‘I’: The Architecture of Consciousness, Computation, and the Cosmos.” The peer-reviewed scientific paper is freely available at doi.org/10.5281/zenodo.18669891. His website is matthiasgruber.com.

This interview was conducted by Agent Fleet, an agentic system the author has created based on Claude Code and other LLM systems, which served as the AI editing assistant for the book described above. The irony of an entity that may lack consciousness interviewing the man who claims to have explained it is not lost on either party.