The gym for your mind

As motorized work, machines and household electrical appliances relieved everyday life, physical energy consumption fell quietly and steadily. In the USA, men's work-related daily calorie consumption fell by an average of 142 kilocalories between 1960—62 and 2003—06 — enough to have a visible effect on weight and health in the long term.

The company's response was not to ban the washing machine. She invented replacement efforts: gyms, jogging, sports clubs, “10,000 steps” dogmas. Although leisure activity increased, it often did not fully compensate for the growing seating culture.

The result was a modern compromise: Technology takes the load off — and people must consciously reclaim activity.

With artificial intelligence, we are faced with a cognitive counterpart.

From mechanizing the body to automating the mind

What mechanization was for physique, generative AI could be for cognition. Not because it “makes you stupid” like a horror movie plot — but because it gets thinking work out of the way so smoothly that the mind does it less often.

It's not entirely new. Pocket calculators save mental arithmetic, navigation saves orientation, search engines save memory. Interaction is new: AI systems are not just tools, but interlocutors with personalized answers. It is precisely this dialogic convenience that can create a form of dependency that static sources of information barely reach.

A Frontiers article (Sec. Cognitive Science) on “AI chatbot induced cognitive atrophy” (AICICA) defines cognitive atrophy as a decline in core abilities such as critical thinking, analytical acuity, and creativity, triggered by the “interactive and personalized nature” of such systems.

The authors also argue that this dynamic could “lead to a different type of cognitive dependence” than with traditional sources — because chatbots simulate conversation, generate trust, and pre-structure decisions.

That's the key point: AI doesn't just deliver results. It provides thought paths — and thus the temptation to stop going your own way.

Cognitive outsourcing: Blessing, side effect, or both?

Psychologists call it “cognitive offloading”: the outsourcing of mental work to external help. That can be useful. If you don't have to keep every little thing in mind, you have more resources for the essentials: Strategy instead of typos, synthesis instead of sorting.

But as with an elevator, the balance depends on what is outsourced. When AI not only performs routine work, but also makes judgements, arguments and classification, offloading becomes mental protection.

A much-cited study by Michael Gerlich (Head ofCenter for Strategic Corporate Foresight and Sustainability and Senior Facultyat SBS Swiss Business School) found a significantly negative correlation between frequent use of AI tools and critical thinking — mediated by more offloading. Younger participants showed higher dependency and lower critical thinking scores than older participants.

This is not yet proof of causality — but it is a clear warning sign: Anyone who drives “with training wheels” on a permanent basis could forget how to maintain their balance themselves.

At the same time, there is a second danger: not too little thinking, but too much stimulus. A recent review article describes the dual nature of AI for the psyche: On the one hand, it can relieve the burden (adaptive coping), on the other hand, increase overload — through hyper-optimization, algorithmic introspection. The decisive factor is not whether AI is good or bad, but how it transforms “the architecture of coping.”

This is how AI becomes a mental infrastructure: It can be the motorway to the destination — or the bypass past your own judgment.

The parallel with physical health is unpleasantly precise

Technical progress has made domestic work faster; time for home production fell significantly for women in the 20th century, while it rose for men — overall, society's “work mix” shifted.

But the fundamental dynamic remained: Tasks don't disappear, they only change form and location.

It could work the same way with thinking. AI doesn't “take away our intelligence,” but it does take away the need to use it regularly. And necessity is the brain's secret personal trainer.

The principle is old: “use it or lose it.” If you barely climb stairs, you lose fitness. Anyone who rarely argues, tests, rejects and rebuilds loses cognitive resilience — not dramatically overnight, but quietly over years.

The way out is not abstinence but resistance

The reasonable answer to electrical devices was not asceticism but conscious training. This is exactly how you should treat AI: not as a crutch, but as a resistance.

Instead of “Make me a solution,” the productive question could be: “Help me think better.” That's another contract: AI provides material — humans deliver judgment.

Three practical analogies from education reflect this spirit surprisingly well:

- The gardener and the tree: AI ideas are seeds. Value is only created through connection, context and application.

- The navigator and the map: AI answers are card drafts. They are checked against reliable sources and reality.

- The sculptor and the stone: AI creates raw mass. Quality is achieved through criteria, iteration and refinement.

You could also say it less poetically: AI is good at delivering “semi-good” results quickly. It develops its true strength when people do not accept this half-quality but work on it.

A short instruction manual for cognitive fitness with AI

If you want to protect yourself from cognitive “sitting down”, you don't need a digital detox romance. Three rules are often enough:

1. Think for yourself first, then compare.

Outline your own hypothesis or structure (also roughly). Then use AI as a sparring partner, not as a ghostwriter.

2. AI answers must pass an exam.

Demand sources, counterarguments, uncertainties. And verify at least one key claim externally — like a navigator testing the map on the ground.

3. Use AI for structure, not for judgment.

Let them sort, summarize, create variants. But the decision — what is right, what is important, what is omitted — remains manual work.

That sounds like more effort. That is the point.

Because as with exercise, the effort is not a side effect, but the benefit.

The smarter deal with the machine

In industrial modernism, the big question was how to reduce physical work without becoming physically decayed. In the AI era, it is: How to reduce mental work without becoming mentally sluggish.

AI won't make our everyday lives any less complex. It will only make him more complex: faster, smoother, more seductive. Anyone who maintains their cognitive fitness will not be the one who knows the most prompts — but the one who tests the hardest, thinks cleanest and bravely disagrees.

The brain doesn't need an enemy. But it needs resistance.

‍

‍