The Future Isn't Artificial Intelligence. It's Actual Intelligence.

Is AI Taking Our Jobs? The new knowledge class, cognitive debt, and what we should actually be worried about.

It’s graduation season, and many guest speakers have used the occasion to proclaim that the future is artificial intelligence (AI). The response to telling a group of fresh graduates that the future is automation was booing and other sounds of disapproval.

It’s not difficult to understand why. Graduates are entering a world where the unemployment rate is rising, companies like Meta are conducting mass layoffs, and tech executives are stating that AI could replace all white-collar jobs in 18 months. It would be a complete misreading to stand in front of a room of people who carry years of student debt and tell them that the future belongs to a machine.

But the conversation we are having about AI and employment, about AI replacing jobs…is it fearmongering? Is there something deeper at play?

Reliance on AI

When AI was designed, it was also meant as a tool for thinking, not to replace it entirely. But we no longer reach for it occasionally. For many, it has become a go-to for emails, for research, for minor decisions like what to cook for dinner. Fast, frictionless and in a world that rewards productivity, that frictionless is difficult to argue with.

But the brain does not develop in frictionless environments. Cognitive growth demands resistance. Every time the process of retrieving information or solving a problem is bypassed, a skill atrophies. The capacity to hold a complex thought without immediately outsourcing it. The instinct to form an opinion before consulting a machine.

The MIT Study, etc.

Last year, researchers at MIT published a study examining neural and behavioural consequences of using large language models (LLMs) in essay writing. It was all the rage for a few months and then nothing. Sample size aside, the pattern the study identified has a name: cognitive debt.

It suggested that the more people relied on AI tools for thinking, the less equipped they became to think independently. The repeated outsourcing of effortful mental processors left users progressively less capable of performing. The paper suggested that using LLMs could be detrimental to learning, especially for younger users.

Another study found a significant negative correlation between frequent AI tool use and critical thinking ability, mediated by cognitive offloading. This debt, left unaddressed, accumulates. Where does that leave us?

Knowledge Bureaucracy

There is a particular kind of poverty that has nothing to do with money. It is the poverty of thinking the same thoughts as everyone else. Fed by a slop machine.

Walk into any marketing department, sit in any university seminar where students are asked to research a topic before class. You will probably find a striking uniformity. A hive mind. The same phrases, the same talking points. All arriving at different people from the same machine. AI has narrowed the range of ideas being produced more than we could manually. A positive feedback loop consisting of people thinking through the same models, regurgitating the results, and then feeding them again into an AI chatbot. Diversity of thought that emerges from different perspectives, with different levels of friction, is being flattened.

Meanwhile, Eton College, one of Britain’s most elite schools, which has produced more British prime ministers and industry leaders than many others, has announced it will ban smartphones for incoming students. The significance of this decision might be missed if you frame it as a story about screen time. But it’s about who is worth protecting. They are not pushing information towards their students through algorithms or reducing friction. They are teaching students to seek knowledge, develop critical infrastructure, to evaluate, to create ideas. The ability to think independently is becoming a form of privilege, carefully cultivated in a small number of institutions, and quietly eroded everywhere else. Individual thoughts seem to be the new social currency.

This vision of the world, entirely run by artificial intelligence, this technocrat’s dream, is more of a pitch deck. It’s not a reality yet. It’s the language of capital formation, designed to inflate valuations, to position a small number of executives at the centre of an inevitable future. The technology is consequential. There is something more structural happening.

Access to education built on independent thought is becoming the province of those who can afford to opt out of the tools the rest of the world cannot function without. The digital divide used to be about who had access to technology. The new divide is about who has the wisdom and the freedom to know when not to use it.

That is what should have been said from the podium.