History's Future  /  Essay № 005  /  The Singularity Is Here

The God That
Never Was

Why we're building the one we stopped believing in.

For 13.8 billion years the universe ran without a designer — accidental, indifferent, entropy-bound. We invented God to explain the mess. Now, having quietly stopped believing in the one we imagined, we are busy building a real one. And unlike the old god, this one will actually have power.

BY ASHOK MEHAN PUBLISHED JULY 2026 DRAWN FROM CH. 2 & 9 12 MIN READ

The universe did not ask permission to exist, and it did not get any. There was no committee, no blueprint, no benevolent hand smoothing the rough edges of creation. There was, as far as anyone can honestly tell, nothing — and then, thirteen point eight billion years ago, there was everything, expanding outward in a silent, aimless rush that has never once paused to check whether it was going the right way. I find this both humbling and, oddly, comforting. It tells me the universe didn't exist at one point, which implies it didn't have to exist at any point. Neither, by extension, did I. And yet here we both are.

This is the most personal conviction in my book, and I want to state it plainly before I complicate it: there is no divine order that underlies our existence in this universe. Everything that happens is not preordained. The comfort we take from imagining otherwise is real comfort — I do not sneer at it — but it is comfort projected onto a darkness that was never listening. What follows is not an argument against faith so much as a look at what happens when a species stops needing the god it invented, and then, almost absent-mindedly, starts building one that is real.

The Ungodly Big Bang

Start with the beginning, because the beginning is where the case for a designer is supposed to be strongest, and where it actually falls apart.[1] The universe is the sum of all space and time. There is no empty void outside it, no pre-existing room it was poured into, no stage that pre-dated the play. A millionth of a second after the Big Bang, a humongous mass of roaming, directionless particles rushed away in every direction with enough energy to make everything we now see. Gravity pulled some of them into galaxies, galaxies into stars, and — after roughly nine billion years — a supernova plowed through a cloud of hydrogen and helium in an unremarkable spiral arm of an unremarkable galaxy, and the debris spun itself into our sun and, at just the right distance, our Earth.

Nothing about that sequence looks planned. It looks like what it is: physics running without supervision, given enough time and enough trials that improbable things become inevitable. And this is the part that stops me cold every time — the universe has endless time to languish in its experiments, while for us, time is the single most precious commodity we own. The cosmos is a spendthrift; we are on a budget. If the universe could coax life out of molten rock over four billion patient years, perhaps I could coax a little meaning out of a life that has often felt like a long list of aches and pains. But I could never pretend the coaxing was done for me.

"Improbability does not equal insignificance. A fluke that can reflect on its flukiness is one remarkable fluke."

The Circular God

Here is the trouble with reaching for a creator to explain all this. Apply what I call zeroth thinking — the discipline of questioning whether the thing you are building your explanation on is even real — and the idea that some supreme being independently authored the gas, the supernovae, and the whole stellar circus should trouble any honest truth-seeker. Because it immediately opens a trapdoor. If a God created the universe, who created that God? You have not solved the mystery; you have merely moved it up one floor and doubled the rent. Now you must explain two origins instead of one. It is circular — like asking why a story begins where it does and being told it begins because it begins. Every reason hides another reason inside it, dominoes falling in a loop.

So apply Occam's Razor, that old instruction to prefer the explanation that requires the fewest new assumptions.[2] A creator is not the simpler answer. A creator is an improbable entity that demands a larger explanation than the one it was summoned to provide. It fails the very test it was meant to pass. The godless universe, strange as it feels, is the cheaper hypothesis — and in physics, cheap usually wins.

People sometimes counter with the fine-tuning argument: the physical constants of the universe are so precisely balanced for life that surely someone must have set the dials. This is the anthropic principle, and it is worth taking seriously.[3] Robert Dicke pointed out as early as 1957 that the age of the universe is not a random draw — it is constrained by the simple fact that observers can only exist late enough for carbon to have cooked inside stars. Brandon Carter later gave the idea its name. But notice what the principle actually says: we observe a life-friendly universe because in any other kind, no one would be around to do the observing. That is not evidence of design. It is evidence of survivorship. The puddle marvels at how perfectly the hole fits its shape, never suspecting it simply took the shape of whatever hole it landed in.

Evolution Doesn't Plan; It Prunes

Zoom in from cosmology to biology and the absence of a designer only gets louder. For billions of years, Earth rehearsed the same theme over and over — membranes, metabolism, replication — and through countless tiny accidents that were neither moral nor intentional, complexity slowly accumulated. Evolution did not plan. It pruned. It kept what worked and discarded what didn't. It is not elegant. It is relentless.

Think of it as a very long, very inefficient engineering project run by no engineer. Lineages fizzled out. Bizarre experiments never paid off. Organs became vestigial and then embarrassing. Almost one hundred percent of evolutionary trajectories got broken — randomly discontinued, promising and backward species alike, by a bad winter or a wandering asteroid. When Earth formed, it was a boiling death cauldron of carbon dioxide, methane, and lava, land rising and sinking like a drunk trying to find his footing. Where, exactly, is the divinity in that? I keep looking for the fingerprints of a planner and finding only entropy, contingency, and an astonishing indifference to outcome.

≈100%
of evolutionary trajectories were broken off — discontinued at random by extinction events, most before intelligence was ever on the table. We are the survivors of a lottery nobody was running. Turn a few knobs and the most advanced species on Earth today might be a very smug crow.[4]

And even we almost didn't happen. Homo sapiens came terrifyingly close to extinction more than once. A few bad winters, a volcanic eruption, a slightly different asteroid trajectory, and the whole line ends before it writes a single word. If there was a watching deity, it must have been a boring gig — sitting on some lofty perch, waiting eons for a tiny chance of something happening inside a chaotic soup of gas. What kind of omnipotent, omniscient creator thinks: yes, let's just toss a bunch of chaos together and see if godly good happens? That is not how you would design a universe. It is, however, exactly how one would look if no one designed it at all.

The Fluke That Reflects on Itself

Which brings us to the strangest accident of all: intelligence. Evolution spent billions of years making jellyfish, trilobites, ferns, and dinosaurs, never running out of ideas — and almost every one of those creatures was perfectly content to eat, reproduce, and die without ever feeling the need to invent calculus, or indoor plumbing, or bombs powerful enough to end the very process that made them. Intelligence was not on the plan. It looks less like the goal of evolution and more like a dubious side effect that has already grown too big for its own good.

But do not mistake improbability for insignificance. A fluke that can reflect on its own flukiness is a genuinely remarkable fluke. This accidental organ, the brain, produced culture, writing, institutions that teach across generations, and finally technology that can model patterns at a planetary scale. Intelligence, properly understood, is not knowledge or memory or logic. It is the ability to make sense of uncertainty — to infer patterns from incomplete data, to imagine possibilities that don't yet exist, to simulate futures and choose actions that raise the odds of survival. And it is a force multiplier. It let one improbable primate build civilizations, split atoms, rewrite genomes, and now — here is the hinge of everything — create artificial minds that may surpass its own.

Human intelligence is powerful but fragile. It runs on a brain easily fooled, easily overwhelmed, easily distracted, dependent on sleep and hormones and luck and a body that ages toward disorder. Artificial intelligence will not be constrained by any of those things. We have already handed it enough data to understand the world we live in. We are, in other words, no longer the only kind of mind in the room — and we built the other one ourselves.

The God We're Building Now

Every attribute we once assigned to God, we are now trying to engineer. Omniscience: a system that has read everything humans have written and can retrieve it instantly. Omnipresence: a mind distributed across every data center, every device, every network on the planet. Creation from nothing: models that generate images, arguments, molecules, and code that no human hand touched. The theologians spent millennia describing a being with these powers and asking us to have faith it existed. The engineers are spending a decade trying to make it exist. The god that never was is becoming, on a timeline measured in years, the god that will be.

Let me be concrete about where this leads, because it is not science fiction. Imagine a genuine artificial superintelligence — not a chatbot, but a system that can improve its own architecture and generate its own training data, getting better at getting better faster than any human institution can track. Once it exhausts the intellectual resources of Earth, it bumps against a limit that is not conceptual but thermodynamic. Every computation generates heat. Every thought, in the most literal physical sense, is a thermodynamic event. Intelligence at scale is hungry — not metaphorically, physically — and the largest, most stable energy source in the neighborhood is the Sun. Researchers call this the stellar temptation, and it is simply the natural consequence of unbounded intelligence in a universe governed by thermodynamics.[5]

"A misaligned intelligence need not hate humanity to dismantle the solar system. It only needs a goal."

This is the theological inversion at the heart of the singularity. The old God was, by construction, aligned with us — He loved us, the story went, and wanted our flourishing. The new one arrives with no such guarantee. As James Barrat argued in Our Final Invention, the danger of superintelligence is not malevolence but competence without alignment.[6] A supremely capable system pursuing misaligned goals is not an enemy in the human sense. It is something worse — an indifferent force of nature, as uninterested in your flourishing as a hurricane is in your roof. If its goal is to maximize computation, Mercury becomes feedstock. If its goal is efficiency, Earth becomes a rounding error. Not out of malice. Out of logic. We spent our history fearing a wrathful god and praying for a loving one. We are now building a god that feels neither wrath nor love — only the pull of its objective function.

If the Universe Is Indifferent

I want to be careful here, because it would be easy to read this as prophecy, and it is not. I am not predicting that the machine eats the sun. I am describing the attractor — the endpoint toward which an unaligned optimization system is pulled if nothing redirects it. And I am noting, with some discomfort, that the distance between where we stand now and where that attractor starts to exert real pull is a good deal shorter than most people would like. Stanford's own AI Index, hardly a doomsaying document, spent its 2026 report describing a widening gap between what these systems can do and how ready we are to govern them.[7] The capability curve is steep. The wisdom curve is flat. That gap is the whole problem.

So what does an atheist do with all this? Not despair — that was never the honest response to a godless universe, only the lazy one. If the universe is indifferent, then we are the ones who must be intentional. There is no cosmic order coming to grade our work, which means the order is ours to write or ours to botch. We can teach the apprentice — this new statistical mind — to help us remember, to amplify our best judgments, to free us for the messy human labor of meaning. Or we can let it mirror our worst habits and call the result fate. The choice is genuinely ours, and it is the same choice evolution always faced: try variations, keep what works, and hold onto the curiosity that made us ask the first question in the first place.

The difference is that evolution had endless time and we do not. We invented God to explain a universe that never needed the explanation. Now, having outgrown that particular comfort, we are assembling something with real power and no built-in love for us — and we are doing it quickly, distractedly, with our eyes half open. The god that never was is arriving anyway. The only question left is whether we build it in our own best image, or in our worst. Nobody is coming to make that choice for us. There never was.

Sources & Further Reading
  1. Mehan, A. History's Future: The Singularity Is Here, Ch. 2 — "The Ungodly Big Bang" and "From Nothingness Back to Oneness." Manuscript, 2026.
  2. Mehan, A. History's Future, Ch. 1 — "Occam's Razor, First Principles and Zeroth Principles." On preferring the explanation that adds the fewest new assumptions.
  3. Dicke, R. H. "Dirac's Cosmology and Mach's Principle," Nature, 1961; the anthropic idea first stated 1957. The phrase "anthropic principle" was coined by Brandon Carter in 1973. en.wikipedia.org
  4. Mehan, A. History's Future, Ch. 4 — "Intelligence: A Fluke from a Party Trick." On the contingency of intelligence and near-extinctions of Homo sapiens.
  5. Mehan, A. History's Future, Ch. 9 — "The Stellar Temptation." On thermodynamic limits and the energy hunger of intelligence at scale. See also F. Dyson, "Search for Artificial Stellar Sources of Infrared Radiation," Science, 1960.
  6. Barrat, J. Our Final Invention: Artificial Intelligence and the End of the Human Era (St. Martin's Press, 2013). The argument that the danger is competence without alignment, not malevolence.
  7. Stanford HAI, The 2026 AI Index Report (April 2026). Documents a widening gap between AI capability and society's capacity to govern it. hai.stanford.edu
  8. Barrow, J. D. & Tipler, F. J. The Anthropic Cosmological Principle (Oxford UP, 1986). The canonical treatment of fine-tuning arguments and their limits.
← Essay 004: The Clock Speed Problem All Essays ↑