Outsourcing Our Minds: How Generative AI May Be Dumbing Us Down

AI tools like ChatGPT promise productivity—but experts warn they could be quietly eroding our critical thinking, memory, and creativity. Are we trading intelligence for convenience?

Author

Elsworth Anthony
April 25, 2025

A Fascination with AI—And a Warning

I’ll be the first to admit it—when I first started exploring generative AI tools like ChatGPT, I was amazed. The ability to generate full essays, answer complex questions, or brainstorm marketing content in seconds felt like science fiction made real. For someone in business and cybersecurity who’s always exploring the latest tools, this felt like a game-changer.

But here’s the thing: I’m also curious about what we might be giving up in return.

A recent Guardian article sent me down a rabbit hole of research, and what I found was both fascinating and unsettling. While AI offers enormous productivity gains, it also raises an uncomfortable question—are we offloading so much thinking to machines that we’re forgetting how to think for ourselves?

This article explores that idea, backed by expert opinions, academic research, and a look at what we need to do next—especially in how we educate ourselves and future generations.

The Age of Cognitive Offloading

“Cognitive offloading” is the practice of using external tools to support or replace internal mental processes. It’s not new—writing things down on paper or using calculators are common examples—but the scale and ease with which AI enables this is unprecedented.

“The more you outsource your cognitive load, the less practice your brain gets at critical thinking and memory retrieval,” says Dr. Anthony Chemero, cognitive scientist and author of Radical Embodied Cognitive Science.

With AI tools capable of drafting emails, summarizing reports, solving math problems, and generating code, we’re seeing a shift from using technology as a support to using it as a substitute.

The danger isn’t just that we’re relying on tools—it’s that we’re no longer challenging our brains. Over time, this lack of challenge can dull the very abilities we used to sharpen through repetition and struggle.

Academic Decline: A Warning Sign?

Educational performance metrics offer tangible signs of this shift. According to the National Center for Education Statistics, student scores in math and reading have fallen dramatically in the U.S. since 2019. In 2024, average reading scores for 8th graders dropped by 4 points from 2022, while math scores plummeted by 9 points for the same group.

The reasons are complex—pandemic disruptions, socioeconomic factors, and teaching quality all play a role. But some researchers are pointing to another culprit: digital saturation.

“We’re in an attention economy, and tech tools—AI included—are designed to minimize effort. But learning is a high-effort activity,” says Dr. Jim Taylor, psychologist and author of Raising Generation Tech.

Multiple studies show that digital natives—those born after 1995—spend less time engaging in deep reading and more time skimming short-form content, leading to surface-level comprehension and declining retention.

The Reversal of the Flynn Effect

For decades, global IQ scores increased steadily thanks to better nutrition, education, and cognitive stimulation—a phenomenon known as the Flynn Effect. But that trend is reversing.

Recent research in Norway, Denmark, and the UK suggests that IQ scores have begun to drop, particularly among younger generations. A 2018 study published in Proceedings of the National Academy of Sciences found that Norwegian IQ scores declined by an average of 7 points per generation since the mid-1990s.

“We’ve replaced many of the mentally stimulating challenges of daily life with automation,” says Dr. Robert Howard, a neuroscientist at University College London. “From GPS to grammar correction, technology handles tasks our brains used to do.”

Generative AI may be the most powerful automation tool yet—and the most subtle in how it alters cognition.

The Misinformation Machine

Aside from its impact on individual cognition, generative AI poses another threat: the erosion of informational integrity.

Large language models like GPT are trained on vast internet datasets that include biased, inaccurate, or false information. While these models are excellent at mimicking human language, they don’t have a built-in truth filter.

A 2023 study from MIT’s Center for Constructive Communication showed that people were more likely to believe information if it was written in fluent, confident language—regardless of whether it was true. AI-generated content, by its nature, is highly fluent.

“AI’s credibility comes not from its accuracy but its tone,” says Dr. Sandra Wachter, Professor of Technology and Regulation at Oxford University. “That’s a dangerous precedent.”

In the wrong hands, generative AI becomes a misinformation weapon—amplifying conspiracy theories, fake news, and fabricated evidence at scale.

Overreliance and the Decline of Skepticism

Perhaps the most subtle risk is how AI erodes our skepticism.

When answers are served instantly and wrapped in well-written prose, we become less likely to question them. This effect, known as automation bias, is well documented in aviation and healthcare—and it’s spreading into everyday life.

“We’re training people to stop thinking critically and start accepting,” warns Dr. Nicholas Carr, author of The Shallows: What the Internet Is Doing to Our Brains. “The more we use AI to explain things to us, the less we feel the need to dig deeper.”

That’s especially problematic in fields like education, journalism, and policy—where depth, context, and rigor are essential.

Reclaiming Human Intelligence: The Role of Education

So, what can we do?

Experts across disciplines agree that the key lies in reforming how we educate and interact with technology. It’s not about rejecting AI—it’s about using it intentionally.

Here are strategies being proposed and piloted globally:

1. AI Literacy in Schools

Teaching students how AI works, its limitations, and how to verify its output should be a baseline skill.

In Finland, students are taught from age 12 how to question online sources, verify facts, and spot AI-generated content. Similar programs are being piloted in Canada, Singapore, and the Netherlands.

2. Cognitive Skill Training

Instead of focusing solely on habit memorization, schools can develop metacognition—teaching students how to think about thinking. Techniques like the Socratic method, scenario-based problem-solving, and interdisciplinary learning encourage analysis over memorization.

3. Active Recall and Deliberate Practice

Psychologists suggest that retrieval-based learning (e.g., flashcards, self-quizzing) is far more effective than passive review. Encouraging students to generate answers before seeking help strengthens memory pathways and analytical reasoning.

AI tools can even support this—by prompting students with questions rather than giving direct answers.

4. Educator Training in AI Integration

Teachers need support too. Professional development programs should empower educators to integrate AI as a co-pilot, not a replacement. AI can assist with lesson planning or provide differentiated learning paths, but educators must remain the guide.

5. Media Literacy Campaigns

Beyond schools, governments and NGOs can run public campaigns to improve critical consumption of AI content. For example, Italy recently launched a nationwide initiative teaching citizens how to identify deepfakes and AI-written propaganda.

Expert Voices: Balancing the Scales

“AI is a mirror—it reflects us. If we approach it lazily, it will amplify that laziness. If we use it rigorously, it can sharpen our capabilities.”
— Dr. Fei-Fei Li, Co-Director, Stanford Human-Centered AI Institute

“The goal should not be to resist AI, but to stay human within it. Critical thinking, empathy, and wisdom are not downloadable traits.”
— Tristan Harris, Center for Humane Technology

“Every generation faces a new cognitive challenge. Ours is to remember how to think in a world designed to think for us.”
— Dr. Maryanne Wolf, Cognitive Neuroscientist and Author of Reader, Come Home

From Hype to Responsibility

Generative AI is not going away. In fact, it’s becoming more embedded in our lives by the day. But with great convenience comes a quiet cost—one that may only become visible when it’s too late.

This isn’t a Luddite warning against innovation. It’s a call for intentionality.

Let’s continue using AI to enhance what we do—but not at the expense of who we are. Intelligence isn’t just about output—it’s about effort, discernment, and creativity. And those are muscles we can’t afford to let atrophy.

Further Reading and Sources