Artificial intelligence (AI) has become the ultimate shortcut in life, for something as simple as deciding what to eat for lunch.
Need an email response drafted in seconds? Ask ChatGPT. Struggling with assignments? Let AI explain. Arguing with a friend about facts, or checking symptoms of your sick child before heading to the hospital… there’s an AI tool for that.
The technology is undoubtedly making life easier, but some researchers are beginning to ask an uncomfortable question: What happens when humans stop thinking for themselves?
Recent studies suggest that as artificial intelligence takes over more of our mental work, people may be remembering less, thinking less critically and relying increasingly on machines to solve problems that our brains once handled naturally.
Experts warn that critical thinking is like a muscle; it grows stronger the more it is exercised through questioning, analysing and solving problems independently. But when AI consistently provides instant answers, that mental workout can become less frequent.
Over time, researchers fear that if people routinely outsource reasoning instead of engaging with ideas themselves, those critical thinking skills could gradually weaken through lack of use, much like a muscle that loses strength when it is no longer exercised.
The concern is not that AI exists, but how we use it.
The rise of 'Cognitive Offloading'
Scientists call it cognitive offloading, the habit of handing over mental tasks to external tools instead of doing the work ourselves.
People have done this before.
The calculator reduced the need for mental arithmetic. GPS replaced paper maps. Search engines made it unnecessary to memorise facts because they were always a Google search away.
Researchers even coined the term "Google effect" to describe how people became less likely to remember information once they knew it could easily be searched online. AI, however, is taking cognitive offloading much further at a very fast and addictive rate.
Instead of helping people find information, large language models such as ChatGPT, Google Gemini and Claude can now write essays, generate business plans, summarise books, answer complex questions and even brainstorm creative ideas in seconds.
According to a recent BBC Future report, researchers are increasingly concerned that outsourcing so much thinking to AI could weaken some of the brain's most important cognitive functions, including memory, creativity and critical thinking.
What MIT Researchers found
One of the most talked-about studies comes from researchers at the MIT Media Lab, led by research scientist Nataliya Kosmyna. The team recruited 54 students and divided them into three groups. One group used ChatGPT to write essays.
Another relied only on traditional internet searches without AI-generated summaries. The third group completed the task using only their own knowledge.
Researchers monitored participants' brain activity using electroencephalography (EEG) while they worked. The results were striking.
Students who relied solely on their own thinking showed widespread brain activity associated with reasoning, creativity and information processing.
Those using search engines still demonstrated strong visual and cognitive engagement.
But participants using ChatGPT recorded significantly lower brain activity, by as much as 55 per cent, according to the researchers.
Kosmyna told the BBC that while the brain did not "fall asleep", there was noticeably less activation in regions responsible for creativity and processing information.
Perhaps even more concerning, many participants who used AI struggled to remember what they had written moments later. Several also reported feeling little sense of ownership over their essays.
Although the research is still awaiting peer review, it adds to a growing body of evidence suggesting that overreliance on AI may reduce mental effort.
The illusion of learning
Memory experts argue that learning happens during struggle. When people wrestle with ideas, compare concepts, organise information and explain it in their own words, the brain forms stronger neural connections.
AI removes much of that productive struggle.
Brain performance coach Jim Kwik believes this creates what neuroscientists describe as an "illusion of learning."
"It feels efficient, but without the effort, you don't get the reward," Kwik explains in a YouTube discussion.
Kwik compares using AI for every task to hiring a personal trainer who lifts the weights on your behalf.
"You wouldn't expect your trainer to do your push-ups for you," he says. "The same applies to your brain."
Instead of replacing thinking, Kwik argues AI should act as a thinking partner.
Young people may be most vulnerable
Researchers are especially concerned about students and children growing up in the AI era, particularly Generation Alpha, those born from around 2010 onwards, who are the first generation to have artificial intelligence integrated into their education and daily lives from an early age.
Several studies suggest that younger people who rely heavily on generative AI may become less likely to question information or solve problems independently.
A study by researchers at the University of Pennsylvania described this behaviour as "cognitive surrender", where users begin accepting AI-generated responses with minimal scrutiny, allowing technology to override their own judgment.
While there is still limited long-term research on Generation Alpha specifically, educators and neuroscientists say the generation could face unique challenges if AI begins replacing rather than supporting learning.
Skills such as memory, curiosity, problem-solving and critical thinking are developed through practice, and experts warn that children who grow accustomed to instant AI-generated answers may have fewer opportunities to strengthen those mental abilities.
The concern extends beyond classrooms. A multinational study found that medical professionals who used an AI tool to help detect colon cancer became less accurate at identifying tumours once the technology was removed, suggesting that overreliance on intelligent systems can affect even highly trained experts in different fields as well.
Researchers say these findings raise important questions about what happens when people begin depending on AI not just for convenience, but for thinking itself.
For Generation Alpha, the challenge may be learning to use AI as a powerful educational tool without allowing it to replace the curiosity, resilience and critical thinking that underpin lifelong learning.
Are we trading Creativity for convenience?
The convenience of AI is undeniable. It can summarise lengthy reports, generate code, translate languages and produce polished writing in seconds. But convenience may come at a cost.
Kosmyna observed that essays produced using ChatGPT often sounded remarkably similar.
Teachers involved in the research reportedly described many of them as "soulless", lacking originality and personal insight.
As more people use identical AI systems, researchers worry that creativity itself could become increasingly homogenised. The danger is not that AI produces poor work. The danger is that humans may stop producing original work altogether.
Researchers have also raised concerns that widespread use of generative AI could make creative work increasingly uniform, with many people relying on similar prompts and producing similar outputs.
In Kenya, the issue recently came into sharp focus after the Copyright Tribunal ruled that works generated by artificial intelligence cannot qualify for copyright protection on their own.
The case involved author Cynthia Beldina Akoth, who had used AI to generate content for Bible Scripture Stories. The tribunal held that while AI can assist in the creative process, copyright protects human originality, meaning creators must be able to demonstrate their own intellectual contribution rather than relying primarily on machine-generated content.
Using AI without losing your mind
Experts stress that AI itself is not the problem. The issue is using it as a replacement rather than an assistant.
Jim Kwik recommends what he calls the ACT Method:
Ask: Use AI to explain concepts, provide structure or outline ideas instead of writing everything for you.
Challenge: Treat AI responses as a starting point. Ask for opposing viewpoints, evidence and weaknesses in its arguments rather than accepting the first answer.
Think: Before typing another prompt, pause and explain the concept in your own words, write a summary from memory or teach it to someone else. Retrieval, not rereading, is what strengthens memory.
Similarly, Vivienne Ming recommends using AI as a debating partner instead of an answer machine.
Rather than asking for solutions, ask the chatbot to challenge your assumptions or poke holes in your reasoning.
That keeps your brain actively engaged instead of passively consuming information.
A future behind a paywall?
Some technology observers believe society could face another challenge in the years ahead.
As AI becomes increasingly embedded in daily life, there are concerns that people may become so dependent on these tools that paying for premium AI services becomes unavoidable.
While there is no evidence that AI companies are intentionally encouraging dependency, many leading models already place their most advanced capabilities behind subscription paywalls.
If people gradually lose confidence in writing, analysing or solving problems independently, the ability to think could become tied to whichever AI service they can afford.
That possibility remains speculative, but it raises an important question about digital dependence and equal access to knowledge.
The people most likely to thrive in the AI era will not necessarily be those who use AI the most, but those who know when to put it aside.
Comments
Sign in with Google to comment, reply, and like comments.
Continue with Google