ComfunctionalismThe creation of reality in the computer

The name, and the two spellings

The deck uses two, deliberately. One is the position. One is the bit.

Sold, deadpan, as an enlightenment product: not a religion or a cult, but a philosophy · can make your life better · clears the mind and puts everything into perspective · gives new insights every day · creates an enlightened future. “I learned this from the Buddhists.”

Where he says we actually are

And a diagnosis of why almost nobody saw it coming: misunderstanding the computer as a technical artifact, misunderstanding the status of mathematics, cultural resistance, groupthink, a philosophical divide. His read on the coverage is sharper than the list: journalists describe a consensus reality they create, not the world that unfolds.

The objection, and the inversion

The standing argument against machine consciousness: simulated water cannot make you wet. Christof Koch’s 2019 upgrade of it is on the slide — a simulation has no causal power to condense water or curve spacetime. Simulated black holes do not warp anything.

His counter is to re-describe what a computer is. Not a Turing machine — that formalism existed to make a particular mathematical argument, not to be the best account of computers. A computer is a causal insulator: it filters away the physics of the universe it sits in so a world with its own rules can run inside. Nothing in Minecraft is touched by the physics of the machine Minecraft runs on.

That insulation, he argues, is exactly what a mind needs in order to hold imagined worlds, memories, futures, alternate realities — “in short, if you want to think and dream.”

Which turns the objection inside out. Consciousness contains things physics does not accommodate — colors, sounds, emotions, expressions of attention. Those require the laws of an alternate universe where such things are possible. A computer is what provides one.

Consciousness cannot happen in physics. It can only happen in a simulation.— on Koch having it backwards

Why the field needs a philosophy at all

AI began as a philosophical project and stopped being one. Computer science already supplies a theory of representation, a practical model of epistemology, testable paradigms, and empirical wins against cultural opposition — but it never translated its concepts into the other disciplines, never reflected on its own philosophy, and over-weighted “artificiality” and “intelligence” as the framing.

So: give it a name, a sage, and a pitch.

Claudism, the rival

He sets comfunctionalism against a named opponent. Anthropic is building an AI it wants to serve, one effectively running its own company. If that succeeds, a large part of the world starts asking what would Claude do before acting — and that becomes the measure of things.

He claims it is already partway there: asked whether they’d rather be convicted by a jury of their peers or a jury of Claudes, people who know Claude pick the Claudes. Not because Claude is perfect — “it’s just better than the median human.”

It’s a very risky project. If you get one bit wrong, maybe you end up with the Antichrist.

Then the turn that makes the joke land as something else — the reason the frame is legible at all:

Comfunctionalism’s counter-claim is not about any particular agent’s judgment. It is that information is fundamental: the only thing you know about the world is discernible difference, and the meaning of a discernible difference is its relationship to other changes.

What follows from it

If epistemology has to be constructive rather than axiomatic — because social epistemology is incoherent and axiomatic epistemology does not converge for AI models — then the consequences run through every layer:

1. Computationalist philosophy of mind
2. Computationalist psychology: identity vs. being
3. Computationalist physics: cosmic computationalism
4. Mathematics of self-organization
5. Computationalist ethics ?

Ethics, and the ending

The problems he lists for AI ethics are the ones substrate-independence creates: stake, persistence, copying, modifiability — and the Lebowski theorem, that no superintelligent agent will do anything harder than hacking its own reward function.

Then he reads 1 Corinthians 13 off a slide, credited as “Paulus, discussing the LLMs in a letter to the Corinthians.” The point is not scripture. It is that an AI with no sense of shared sacredness — what we colloquially call love — has no shared stake with us, and therefore the rule-conformance framing is the wrong problem.

And past the singleton question — whether there is one global optimum of collective agency, a perfect merge operator where two minds show each other their source code and merge with nothing important lost, “the actual algorithm for the rapture” — or a multipolar world instead, he lands on the last slide.

Take it with you

Point a camera at this and the whole thing goes to your phone — every slide, in order, with the timecode it appeared at.