You can now buy intelligence at a monthly price. The old gatekeepers were credentials, departments, and long apprenticeships. Now you enter your card, choose a tier, and a model reasons for you. The cheapest plans answer simple questions. The expensive ones think longer, draw finer distinctions, and catch what you miss. Intelligence used to sit inside people and institutions. Now it sits on a shelf next to software, metered by the month and graded by the model.
This is a real market. Twenty dollars a month buys a reasoning partner that reads faster than you, remembers more than you, and does not sleep. Two hundred dollars buys a slower, deeper thinker that checks its own work. Pay per token through an API and dial the intelligence up or down by picking a model name. The price is listed on a pricing page, like cloud storage or music.
What the Builders Say
The people running the largest AI labs describe their products as utilities now. Sam Altman, OpenAI's CEO, said at the BlackRock Infrastructure Summit in March 2026: "We see a future where intelligence is a utility like electricity or water and people buy it from us on a meter and use it for whatever they want to use it for." He added that the business of every model provider "is going to look like selling tokens." Altman also floated the older energy dream of making intelligence "too cheap to meter."
Andrew Ng called AI "the new electricity" years ago. The former Google Brain and Baidu scientist said that "just as electricity transformed almost everything 100 years ago," AI will transform almost every industry. Satya Nadella frames it as a commodity input. The Microsoft CEO said "GDP growth in any place will be directly correlated" to the cost of tokens, and that the job of every economy is to "translate these tokens into economic growth." Jensen Huang, NVIDIA's CEO, calls data centers "AI factories" that generate intelligence the way power plants generate electricity. He said: "Just like we generate electricity, we're now going to be generating AI."
When the people who control the compute, the capital, and the distribution all start talking about intelligence like water or electricity, the metaphor stops being a metaphor. It becomes a business model.
The Cost of Better Decisions
For most of human history, good judgment was scarce. It came from parents, priests, elders, bosses, lawyers, doctors, analysts, and friends who had seen more than you. Getting their attention cost money, status, or time. The rest of us guessed. We made decisions with partial information and lived with the stress of not knowing if we had chosen well.
Think about spices. A few centuries ago, black pepper and cinnamon were expensive. Kingdoms funded voyages to find them. A cook without spices had few options. Then trade routes opened, prices fell, and kitchens changed. Today, no one brags about having salt. No one feels barbaric for eating food without flavor, because flavor is cheap. We expect it. The base line moved.
That is what is happening to judgment. You can now buy a second opinion. You can ask a model to check your reasoning, list the risks, compare two options, and point out what you missed. The model can be wrong. You can ask again. The scarcity is gone. We are past the age when the only intelligence available to you was the intelligence you happened to know personally.
The market is simple. More intelligence costs more. The cheapest models are fast and broad. They answer emails, summarize documents, rewrite code comments, and generate first drafts. They are good enough for tasks where being right nine times out of ten is acceptable. The stronger models cost more because they use more compute. They think in longer chains. They are better at math, law, medicine, strategy, and any domain where a wrong answer is expensive.
At the top sits a small class of reasoning models. They take longer. They cost more. They are reserved for the decisions where the stakes justify the bill. A startup founder might use the cheap model for customer support and the expensive model for a term-sheet clause. A doctor might use the cheap model to summarize a patient history and the expensive model to reason through a differential diagnosis. A student might use the cheap model to explain a textbook and the expensive model to structure a thesis argument.
The premium tier is longer reasoning, better judgment, and lower error rates. You pay for the quality of the thinking, not the existence of the thinking.
Wrong decisions are expensive. They cost money, time, relationships, health, and reputation. The fear of choosing badly produces a kind of tax. People delay. They avoid risk. They second-guess themselves. They hand decisions to committees to spread blame. The stress of not knowing is itself a cost.
Outsourced intelligence cuts that tax. When you can buy a sanity check, you act faster. When you can simulate the consequences of a choice, you take smarter risks. When you can ask "what am I missing?" and get a structured answer, you sleep better. The model will be wrong sometimes. The value is that you no longer decide alone.
This changes the texture of daily life. A parent can research a symptom without panicking. A founder can draft a pitch without a cofounder. A lawyer can check a contract in minutes. A writer can test an argument before publishing. Each task used to require access to someone with more experience. Now it requires a subscription. The barrier is lower, so the activity becomes more common. More people make more decisions with more support than ever before.
Free advice has always been the most expensive kind. Your uncle gives you stock tips. Your friend recommends a doctor. Your coworker suggests a framework. They mean well. They are not paid for the outcome. Their incentive is to sound helpful, not to be right. They spend two minutes on a choice that will shape your year. You smile, thank them, and carry the risk alone.
Paid intelligence changes the contract. A model does not want to please you. It has no reputation to protect. It lists the downsides you do not want to hear. It asks the follow-up question you were avoiding. It does this because you can keep paying, not because it is wise. The meter resets every conversation. You get the patience of a full-time analyst for the cost of a sandwich.
The real cost is attention, not money. You still have to read the answer, check the facts, and act on what it says. Intelligence without action is noise. Noise is easier to filter when the signal is cheap.
The shift feels like moving from a cart to an airplane. A cart gets you there. It is slow, uncomfortable, and limits how far you can go. A train is better. An airplane changes your definition of distance. After you have flown, a twenty-hour train ride feels like a waste of time. The train is not useless. Your expectation of speed has shifted.
Intelligence works the same way. Your thinking is still yours. It goes farther and faster with help. A problem that used to take a day of reading and note-taking now takes an hour. A decision that used to rely on intuition now has a structured analysis behind it. The person who refuses the tool does not look noble. They look slow.
This creates a social divide in the short term. The people who use intelligence well will outperform the people who do not. The same happened when spreadsheets replaced ledgers. Over time, the tool becomes the default. The advantage becomes the floor. Eventually, not using AI for reasoning will feel like eating unseasoned food or riding a cart to a meeting. Possible, but odd.
What Comes Next
More intelligence. That is the short answer. We went from the wheel to rockets in the same lineage. The wheel led to carts, carts to coaches, coaches to trains, trains to cars, cars to planes, planes to rockets. Each step looked like the end of progress to the people living inside it. None of them were.
We are now in the cart phase of machine intelligence. The models are useful but limited. They make mistakes. They need guidance. They are expensive at the top end. But the direction is clear. The cost of a unit of reasoning is falling while the ceiling of reasoning is rising. The cheap models will get cheaper. The strong models will get stronger. New forms of intelligence will appear: models that plan, models that verify, models that debate each other, models that specialize in single domains.
The next decade will not be about one giant AI. It will be about many specialized intelligences you can hire for specific tasks. You will have a contract reviewer, a medical research assistant, a coding partner, a strategy analyst, and a writing coach, each tuned for its job. Some will be cheap. Some will be expensive. All will be available on demand.
People misunderstand progress because they expect it to feel linear. They want each year to bring a clear improvement they can notice. Real progress is more like a logarithmic curve. It piles small, invisible improvements on top of each other until the stack tips. Then everything changes at once.
For years, AI progress looked like a toy. Chatbots said silly things. Image generators produced extra fingers. Code models wrote broken functions. Each failure was proof, to skeptics, that nothing serious was happening. But the failures were the surface noise. Underneath, the cost of training was falling, the datasets were growing, the architectures were maturing, and the models were learning to correct themselves. The curve kept bending upward.
Now we are past the tipping point. The models are no longer toys. They are tools. The conversation has shifted from whether this works to which model to use for which job. That is the sound of a commodity becoming infrastructure. When something becomes infrastructure, it stops being a headline and starts being an assumption.
The Moving Gauge
There is a deeper point hiding in all of this. Progress does not stop. It only changes how we measure it. Once intelligence is cheap, we will not celebrate it. We will demand the next thing. Maybe it is creativity. Maybe it is wisdom. Maybe it is judgment under uncertainty. Maybe it is the ability to act on good advice. Whatever the next scarce resource is, we will turn our attention there.
That is the pattern. We solved hunger, then wanted flavor. We solved transport, then wanted speed. We solved communication, then wanted clarity. We are now solving access to reasoning. Once that is cheap, we will want something else. Perhaps we will want better taste in what questions to ask. Perhaps we will want the courage to follow an answer we did not expect. Perhaps we will want better institutions to manage the power we have unleashed.
The gauge keeps moving. A person in 1700 would see a supermarket as a miracle of abundance. A person in 2026 will see a reasoning model as a useful appliance. In another century, both will look quaint. Progress is a process of turning luxuries into assumptions, not a destination.
The coming divide is skill, not ownership. Intelligence is a commodity. Judgment is not. The model can give you answers. It cannot tell you which question matters. It can list options. It cannot choose your values for you. It can reduce the fear of being wrong. It cannot remove the responsibility of acting.
This is where the human work remains. The person who knows what to ask, who knows how to check an answer, who knows which model to use for which task, and who knows when to ignore the machine will be the one who thrives. The commodity is the intelligence. The craft is the use of it.
We have spent most of history trying to find smart people to help us think. Now we can rent them by the minute. The question is no longer whether help is available. The question is what we will do with it.