Does AI Make Us Dumber? 3 Studies and the Real Risk

Quick summary: does AI make us dumber? Not in a biological sense, but as AI takes over more of our thinking — remembering, calculating, writing, deciding — we risk a form of cognitive dependence: not evolutionary regression, but a gradual weakening of skills we simply stop practicing. The fix isn’t rejecting AI, but using it as a partner while deliberately continuing to exercise human judgment, creativity, and reasoning.

For millions of years, our human ancestors survived because they solved problems. They hunted, navigated unfamiliar landscapes, remembered water sources, built tools, developed languages, and created civilizations. Every challenge demanded thought, creativity, memory, and adaptation. The human brain became the most sophisticated biological computer on Earth because it was continuously exercised by necessity.

Today, for the first time in history, humanity is entering an era where technology is beginning to think on our behalf.

Artificial Intelligence can write articles, solve mathematical problems, generate computer programs, create artwork, translate languages, diagnose diseases, and even make strategic decisions. While this technological revolution promises unprecedented productivity, it also raises a profound scientific question:

If machines increasingly perform our thinking, what happens to the human brain?

Key Takeaways

As AI takes over more remembering, calculating, writing and deciding, humans risk cognitive dependence — a gradual weakening of skills we stop practicing, not evolutionary regression.
Early research, including an MIT Media Lab preprint and a Microsoft-Carnegie Mellon survey, hints that heavy reliance on AI can reduce mental effort, though the evidence is still preliminary.
The fix isn’t rejecting AI — it’s using it as a partner while deliberately continuing to exercise human judgment, creativity, and independent reasoning.

Evolution Rewarded Thinking

Evolution does not reward intelligence for its own sake. It rewards survival.

Throughout the evolution of human ancestors over millions of years—and especially during the roughly 300,000 years of modern humans—the brain expanded and became increasingly sophisticated because those with better planning, communication, memory, and problem-solving abilities were more likely to survive and reproduce.

Every generation faced new challenges.

Finding food

Escaping predators

Understanding nature

Building communities

Inventing tools

Developing agriculture

Creating science

The brain was constantly being trained by real-life complexity.

The Modern Brain Faces a Different Environment

Today’s environment is fundamentally different.

Instead of remembering directions, we rely on GPS.

Instead of performing calculations, we use calculators.

Instead of memorizing information, we search online.

Instead of writing from our own thoughts, many people ask AI to generate content.

Technology has become an extension of the human brain.

This phenomenon is known in cognitive science as cognitive offloading—the tendency to rely on external tools to reduce mental effort. While this improves efficiency, excessive dependence may reduce the regular practice of memory, attention, reasoning, and independent problem-solving.

Like muscles, cognitive abilities improve with use and may weaken when they are rarely exercised.

Diagram showing how cognitive offloading to AI reduces deep problem-solving endurance, the main way AI can make us dumber — The cognitive offloading loop: convenience leading to reduced analytical stamina.

The “Use It or Lose It” Principle

Neuroscience shows that the brain possesses remarkable plasticity.

Neural connections strengthen through repeated use and weaken when neglected.

This principle is well established in research involving:

Memory

Language learning

Mathematical reasoning

Motor skills

Musical training

When technology consistently replaces mental effort, the brain receives fewer opportunities to build and reinforce these networks.

The concern is not that the human brain will suddenly become biologically smaller, but that our everyday cognitive performance could gradually decline if we stop exercising key mental abilities.

Does AI Make Us Dumber? What the Early Research Says

Does AI make us dumber? The honest answer is that nobody knows yet, because tools like ChatGPT are only a few years old. But a few early studies point in the same direction as the concern in this essay. Three are worth knowing, along with their limits.

The “Google effect” (2011). Psychologist Betsy Sparrow and colleagues found that people were better at remembering where to find information online than the information itself.
A Microsoft and Carnegie Mellon survey (2025). It asked 319 knowledge workers who use generative AI at least weekly how they apply critical thinking. Its authors reported that confidence in AI was associated with reduced critical thinking effort, while workers confident in their own skills were more likely to scrutinise the output. The results are self-reported, and The Register notes that six of the seven researchers work for Microsoft.
MIT Media Lab, “Your Brain on ChatGPT” (2025). In this study of 54 participants, people wrote essays using an AI chatbot, a search engine or no tools while EEG measured brain activity. The chatbot group showed the weakest brain connectivity, had difficulty quoting their own essays and reported the lowest sense of ownership. It is an arXiv preprint with a small sample.

None of this proves that AI damages the brain. These studies are small, early or self-reported, and lower brain activity during a task can also mean the task became easier. What they do suggest is the pattern this essay describes: skills we stop practising may fade. A calculator did not make humanity unable to think, but a student who never learns arithmetic may struggle without one.

Intelligence Is More Than Information

AI has access to vast amounts of information.

Humans possess something different.

Wisdom

Moral judgment

Empathy

Creativity

Intuition

Ethical reasoning

Emotional intelligence

Social understanding

These qualities arise from lived experiences, relationships, emotions, and values—not simply from data.

A civilization that delegates every decision to algorithms risks becoming efficient but less thoughtful.

The Education System Faces a Historic Challenge

For centuries, education focused on teaching students how to solve problems.

If AI solves every assignment, writes every essay, and answers every question, students may begin learning how to use AI instead of how to think independently.

The danger is not AI itself.

The danger is becoming passive consumers of intelligence rather than active creators of knowledge.

Schools and universities must therefore shift their focus toward:

Critical thinking

Scientific reasoning

Creativity

Debate

Practical experimentation

Ethical decision-making

Interdisciplinary learning

These are areas where human cognition remains indispensable.

The AI Paradox: Will Human Evolution Reverse?

Some people fear that AI could reverse the evolution of the human brain.

From a scientific perspective, this claim remains speculative.

Biological evolution occurs over many generations through changes in reproductive success. There is currently no evidence that AI is causing humans to evolve genetically toward smaller or less capable brains.

However, cultural evolution happens much faster than biological evolution.

If future generations rarely practice memory, calculation, navigation, writing, or deep reasoning because technology performs these tasks, society could experience a decline in important cognitive skills—even if our biology remains unchanged.

In other words, the greater risk is not evolutionary regression but cognitive dependence.

The Mental Health Dimension

Continuous dependence on digital technology also influences attention and behavior.

Researchers are increasingly studying links between excessive digital engagement and:

Reduced attention spans

Information overload

Decision fatigue

Increased anxiety

Digital addiction

Reduced patience for deep work

AI may amplify these trends if people become accustomed to receiving immediate answers instead of engaging in prolonged thinking.

A Civilization That Stops Thinking

History shows that civilizations advanced because people questioned accepted beliefs.

Every major breakthrough—from the wheel to electricity, from vaccines to space exploration—was born from curiosity, persistence, and human imagination.

If future generations become satisfied with machine-generated answers instead of asking deeper questions, innovation itself could slow.

Progress requires questioning.

Machines provide answers.

Civilizations grow because humans ask better questions.

The Right Relationship with AI

The objective should never be to reject Artificial Intelligence.

The objective should be to use AI as a partner rather than a substitute.

AI should handle repetitive tasks, process large datasets, and enhance productivity.

Humans should continue to exercise judgment, creativity, ethical reasoning, imagination, and original thought.

Technology should amplify human intelligence—not replace it.

How Can You Use AI Without Losing Your Own Thinking?

These are practical habits that follow from the idea of AI as a partner, not scientific prescriptions:

Think first, ask second. Write your own answer or outline before asking AI, then compare.
Use AI to check, not to replace. Ask it to find gaps in your draft instead of writing the draft for you.
Verify what it tells you. AI can sound confident and still be wrong, so keep the habit of checking important facts.
Keep some tasks AI-free. Mental arithmetic, remembering a route or writing a letter by hand work like exercise for the mind. Simple daily routines, such as the ones in Master Your Mind Before You Master Your Life, also help.
Ask children to explain answers in their own words. Patience for slow thinking is a skill, and the impatience that instant answers create is easier to prevent than to cure.

Continuous learning is also the best career protection in the AI age; as I argued in Your Job Is Not Your Career, the strategy that made you successful at 25 may not be enough at 45.

Final Thoughts

Artificial Intelligence may become one of humanity’s greatest inventions.

Yet history teaches us that every powerful technology transforms the people who use it.

The greatest challenge of the AI age may not be building machines that think like humans.

It may be ensuring that humans never stop thinking for themselves.

This is the AI paradox in its purest form: the future of civilization will not be determined by how intelligent our machines become.

The AI paradox will be decided by whether human intelligence continues to evolve alongside them.

Frequently Asked Questions

What is the AI paradox?

The AI paradox is the tension between building increasingly intelligent machines while potentially making our own thinking less active. As AI performs more remembering, calculating, and writing on our behalf, everyday human cognitive skills get fewer chances to be exercised.

Is AI reversing human evolution?

No — from a scientific perspective this remains speculative, since biological evolution happens over many generations through reproductive success, and there is no evidence AI is causing genetic change. The real risk is cultural and behavioral: cognitive dependence, where skills weaken from disuse, not a biological reversal.

How can people avoid losing cognitive skills to AI?

The article’s answer is not to reject AI but to treat it as a partner rather than a substitute — letting AI handle repetitive tasks while deliberately continuing to exercise judgment, creativity, memory, and reasoning. Neuroscience’s “use it or lose it” principle means these abilities strengthen with regular practice and weaken when neglected.