Sometime in the last two years the price of intelligence fell by a factor of two hundred and eighty, and almost nobody noticed.
In November 2022 it cost about twenty dollars to get a million tokens of GPT-3.5-level reasoning out of a machine. By October 2024 the same quality of thinking cost seven cents. A 280-fold collapse in the price of the thing our entire civilization is supposedly built on, in twenty-three months. 1 If the price of oil had done that, it would have been the only story on Earth.
It wasn’t. It was barely a story at all. And the reason it wasn’t, I have come to think, is the same reason I have come to believe the future is going to be bright.
I have never understood cyberpunk. Not the aesthetic. The aesthetic is fine. I like rain and neon as much as anyone. The premise. The premise is always the same: breathtaking technology, flying cars, brain implants, holograms in the street, and underneath it a ruined planet, a starving underclass, and three megacorporations running everything. Half the future is turned up to eleven and the other half is left to rot.
It is not a prediction. It is a contradiction wearing a prediction’s clothes.
The argument I want to make about that comes from David Deutsch. Problems are inevitable. Problems are soluble. The human ability to transform nature is limited only by the laws of physics. Anything not forbidden by physics is achievable, given the right knowledge. The only thing standing between us and any particular transformation is knowing how.
Apply that to the flying car. To build one that ordinary people use, you need four things. Energy so cheap that lifting a ton of metal into the air all day is economically trivial. Materials light and strong enough that the thing flies at all. Artificial intelligence good enough to manage millions of autonomous vehicles in shared airspace without killing anyone. And an economy rich enough to pay for the other three.
But look at what each of those four things also does, the moment you have it. Cheap energy is the entire problem of carbon removal. Pulling CO₂ out of the atmosphere is not forbidden by physics. It is expensive, and expensive is a synonym for needs cheap energy. A civilization with flying-car energy prices removes its carbon backlog the way we pave roads. Boring, scheduled, unremarkable.
Materials science good enough for flying cars is materials science good enough for cheap desalination, engineered crops, and ecological restoration at scale. The AI that safely flies a million cars is the AI that runs a personal doctor and a personal tutor for every human being. And the economic surplus that pays for all of it is, by definition, the opposite of the starving underclass the story insists on.
So the cyberpunk future asks you to believe something genuinely strange. That a civilization solved the hard problem, limitless clean energy, advanced AI, abundant materials, and then sat in the rubble for a century declining to solve the easy ones that the hard solutions hand it for free. That isn’t dark. It is incoherent. It is a society that built the ladder and then refused, on principle, to climb the last rung.
The honest extrapolation runs the other way. If you genuinely believe in the flying cars, you are already committed to the regreened Earth. They come from the same knowledge. The solarpunk future, clean and abundant and alive, is not a softer alternative to the cyberpunk one. It is the cyberpunk one with the arithmetic finished.
The doom is the part that doesn’t follow. When I say the optimistic case is the logical one, people assume I’m doing vibes. So here is the opposite of vibes. Three things that happened in 2025, and a note on the address.
The first AI-designed drug reached patients. A treatment for idiopathic pulmonary fibrosis, a disease that for as long as we have had a name for it has killed people on a steady schedule, where the machine did both halves of the work: picking the target and designing the molecule. The lab is in Hong Kong. The Phase IIa results were published in New York.
A second cure moved from clinic to clinic. By late 2025 nearly three hundred sickle-cell patients had been referred for the first gene-editing medicine ever approved. Sickle cell, a thing that for the entire history of medicine you simply had, is now, for some people, a thing you had and no longer have. The science came out of Cambridge, Massachusetts and Switzerland.
And we started leaving the planet at a cadence that looks like a typo. Three hundred and twenty-nine orbital launches in a single year, a 25% jump from the year before. One company flew more than half of them by itself. That company is in Texas.
Notice the addresses. The molecule, Hong Kong and New York. The cure, Massachusetts and Basel. The launches, Texas. The frontier models, Mountain View and South San Francisco. The humanoid robots finally doing real work on real assembly lines, Sunnyvale and Shenzhen. The continent I live on barely appears in this list yet, and that is the part I want to change.
One of the reasons I think it is changeable is the company I work at. Helsing is a defense AI company in Munich, and one of the most valuable private companies in Europe right now. It exists because a small number of people in Europe looked at the speed of the room they were sitting in and decided it was unacceptable, and built a different room. There is no law of physics or economics that says there cannot be a hundred more of them. There is just a decision, taken one founder at a time, to stop watching and start building.
That is the mechanism, wherever the bright future is being made.
Here is the part that should be reassuring and somehow never is. None of this is new. The curve has been bending the right way for two hundred years. We are simply built not to notice.
In 1820, three out of four humans lived in extreme poverty and one in eight could read. Today extreme poverty is around one in ten and most adults on Earth read. In 1800, forty-three of every hundred children died before their fifth birthday. In 2024 it was under four. Global life expectancy has gone from the high twenties to over seventy-three and shrugged off a global pandemic in two years.
Underneath all of it, the cost curves that make every story above possible kept doing the one thing the pessimist’s model says is impossible. They kept collapsing.
Solar modules are 99.8% cheaper than they were in 1975. Battery packs have fallen 93% in fifteen years. Sequencing a genome cost ninety-five million dollars in 2001 and a couple hundred today, a decline that makes Moore’s Law look lazy. In 2025, for the first time since 1919, renewables generated more of the world’s electricity than coal.
You can dislike any one of these. You cannot look at all of them at once and tell me the machine is winding down.
So why does it feel like the opposite? Because progress is gradual and disasters are sudden, and our brains were tuned for a world where the sudden thing was the one that ate you. A plane crash springs to mind. Four hundred thousand children not dying this year springs to mind for no one, because non-events do not have photographs.
This is the deepest reason I distrust pessimism. It isn’t that pessimists have bad data. It is that the pessimistic feeling is manufactured by the structure of attention itself, and then mistaken for insight. The cleanest demonstration is an old one. In 1980 an economist bet the most famous doom author of the era that the price of any five metals he wanted would fall in real terms over the next decade, because as the population grew the inventive capacity would grow faster. Ten years later every single one had fallen, the basket down 36%. The doom author mailed the economist a check. 2 More people had meant more minds, and more minds had meant more ways to do more with less. That is the whole optimist thesis in one wager.
I want to be careful here, because there is a lazy version of everything I have just said, and it is worth naming so I can disown it.
The lazy version is: technology will fix it, so relax. Sit back, the curves go up on their own, the future arrives on schedule like a train. That version is wrong, and it is wrong for exactly the reason I wrote about in Hormesis. What doesn’t get loaded doesn’t get strong. Bone in zero gravity dissolves, not because space is cruel but because it is too kind, because nothing is asking anything of the bone. A future nobody works for is bone in zero gravity. It doesn’t arrive soft and pleasant. It just doesn’t arrive.
None of the numbers in this essay happened on their own. There is no law of nature that produced 329 launches in 2025. There was one company that refused to accept that launch was expensive, and a man who bet everything on it more than once. There is no celestial schedule that delivered a gene-editing cure. There were specific scientists who refused the timeline. A tool only multiplies people who pick it up and push harder, not the ones who use it as a reason to coast.
Hormesis was about individuals: that you are built by the loads you choose to carry. This is the same claim at the scale of a civilization. The good future is the effortful one. It is earned by a relatively small number of people working on hard things, unfashionably, for a long time, against a current that is forever telling them to stop.
And the current is getting stronger. Negativity used to be a mood. Increasingly it is a program, and there are three currents in it worth naming, because vague gestures at the doomers let the actual arguments off the hook.
The first is the proposal to deliberately shrink the productive economy. To consume less, produce less, build less, on the theory that the responsible response to environmental pressure is to step on the brake. This is no longer a fringe position. There was a conference inside the European Parliament a few years ago titled Beyond Growth, opened by the President of the Parliament and the President of the Commission. People are arguing, in the actual chambers where the actual laws get made, that the cure for our problems is to do less of the thing that produces the wealth and the knowledge our problems require. If problems are soluble, and solving them requires knowledge, and knowledge grows out of a thriving, productive civilization, then deliberately shrinking that civilization is not caution. It is the patient lying still in the bed because moving hurts, while the muscle wastes. It is the zero-gravity move, dressed up as virtue.
The second is demographic surrender. Most countries on Earth will be below replacement fertility before the end of this century, and a striking amount of elite commentary treats this as good news. Fewer humans, lighter footprint, problem solved. Every name in the optimistic half of this essay was a person. People are the ultimate resource in the most literal possible way. A civilization that decides, on principle, to produce fewer minds is a civilization that has decided to solve fewer problems. You cannot be a Deutschian about physics and a Malthusian about people. The minds are where the solutions come from.
The third is harder to name. A generation has been raised on the conviction that the world is ending. Not as a worry. As a setting. By twelve they have been told the planet has a decade left. By eighteen they have been told they were lucky to be born, but luckier still not to have children of their own. They arrive in their twenties already exhausted, and the exhaustion gets read, by them and by everyone above them, as wisdom.
It is not wisdom. It is miscalibration.
Climate change is the cleanest example. It is, strictly speaking, a technical problem with a known solution path. You need cheap clean energy, materials capable of carbon removal at scale, and the economic surplus to pay for both. Every one of those is moving in the right direction on a curve steep enough to be visible from orbit. None of that guarantees we get there in time. All of it makes the metaphysical version, the we are the last generation, the oceans will boil, there is no point in trying, a category error. A soluble engineering problem is being lived as an extinction event.
I think the reason a generation cannot tell the two apart is the one I worried about in Hormesis. Nobody let them lose a small game. The first real challenge of their lives arrived prepackaged as the end of the world, because that is how the room they grew up in described every challenge. They have no calibration for the difference between a problem that is hard and a problem that is hopeless, because every problem they were handed came pre-labelled hopeless. The participation trophy and the climate documentary are the same intervention, applied to opposite ends of the emotional range, with the same result: a person who cannot tell a load from a disaster.
To be clear, so the lazy reading has nowhere to hide. The climate is warming. The arctic is melting. Species are dying. None of that is in dispute. What is in dispute is the response. The response of the people in this essay who are actually working on it is to build, fast, hard, in the direction of the solution. The response of the people who have decided to be sad about it is to be sad about it. Only one of these is going to do anything to the temperature of the planet.
The systems building the bright future fastest are the ones that can change direction. Texas can decide to build a launch site and build one. California can decide to fund a trial and fund one. A small startup in Munich can decide to design a piece of defense software a continent’s worth of generals said was impossible, and ship it anyway. None of these are coordinated. None of them are central plans. They are a thousand small refusals of the speed of the room.
This is not really a sentence about America being better and Europe being worse. It is a sentence about a kind of person, who exists on every continent and inside every system, and who is responsible for almost everything in this essay. Europe has produced that kind of person before. It is producing some of them now. It needs to produce many more, and it needs to stop telling itself that prudence is a substitute for building. The bright future is real, and it is permitted by physics, and it is also permitted by Europe. The only question is how many of us decide to go and get it.
Let me give the other side its strongest version, because I would want one if I were on it. Some serious people think advanced AI is a genuine existential risk rather than a productivity story, and they are not obviously wrong. A great-power war could shatter the supply chains every curve in this essay quietly depends on. A pandemic worse than the last one could arrive before our defenses are ready. Optimism that ignores dose is just the lazy version in a different costume. So I am not promising the good future. Nobody can. What I am rejecting is the claim that the bad future is the sophisticated, clear-eyed, grown-up expectation, and optimism the naive one. It is the reverse. The dystopia is the incoherent story. The decline is the artifact of how attention works. The arithmetic, the actual numbers from this actual year, all point the other way. Toward a world with cheap clean energy, cured diseases, a personal tutor and a personal doctor for every human being, robots for the dull work, and a road back to the Moon and outward. That world is permitted by physics. The only question is whether enough of us decide to go and get it.
Being alive right now is itself the answer to the question of whether to be excited. We are alive in the years the cost of intelligence fell by two hundred and eighty times, and the first edited gene cured the first person, and the first city’s worth of rockets left the planet in a single year. If you cannot find that thrilling, the problem is not the future.
None of it is guaranteed. That is not the bad news. That is the assignment. The reason a child born this year will live in a world with fusion power and longevity drugs and a thriving green Earth is if and only if enough of us spend the next thirty years working as though it depends on us. Because it does.
So put down the cyberpunk novel. The arithmetic is on our side, the century is wide open, and the photograph nobody is taking is the one with you in it.
The 280-fold figure is the AI Index headline number against a particular GPT-3.5-equivalent benchmark. Reasonable alternative measures of the same window produce numbers between 50x and 500x. The point is the order of magnitude, not the decimal. The right way to read it is that a curve almost everyone would have called impossible ten years ago is now routine, and the routine is what nobody photographs.
↩︎Julian Simon and Paul Ehrlich, 1980. Ehrlich picked metals, which Simon argued was the worst class of resources for Ehrlich’s thesis, and subsequent academic redos with different baskets and decades produce mixed results. What survives all of them is the structural intuition: more people means more minds means more substitutes. The bet didn’t prove population can never strain anything. It demonstrated that betting against the inventive capacity of humanity is a bad bet over any horizon long enough for the inventing to happen.
[↩︎](#fnref:2)
[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)by the author.