/*/*]]>*/ When AI Thinks for Us: The Cost of Cognitive Atrophy - LapComp

When AI Thinks for Us: The Cost of Cognitive Atrophy

Illustration of a human silhouette interacting with digital and physical learning elements representing cognitive balance

We live in an era of unprecedented digital assistance. Recently, scholars like MIT Professor Eric So have started turning a critical lens inward, questioning what happens to our own minds when algorithms do the heavy lifting. In his book, The Collision: What AI Does to Us, So examines the friction points between human cognition and widespread automation. While these tools undeniably speed up our daily workflows, they also invite us to question whether convenience is quietly eroding our mental agility.

The Allure of the Shortcut

It is remarkably easy to fall in love with frictionless productivity. When a complex email, a dense coding error, or a messy financial spreadsheet can be cleaned up in seconds by a large language model, the temptation to delegate is overwhelming. After all, why struggle through a cognitive bottleneck when a machine can offer a polished solution instantly?

Yet, looking at this through the lens of human psychology, we run into a fundamental truth about how the brain actually develops. Cognitive growth has never been about reaching the final answer as fast as possible. Instead, it is the messy, uncomfortable middle part—the friction of wrestling with material, failing repeatedly, and iterating—that builds durable neural pathways.

Understanding Cognitive Atrophy

In my view, the most compelling argument raised by critics of hyper-automation is the concept of cognitive atrophy. Just as a muscle weakens when it is relieved of physical resistance, our capacity for deep critical analysis, synthesis, and long-form memory appears to degrade when we consistently outsource our thinking to machines.

Think of it like relying entirely on GPS navigation. While you will eventually reach your destination without getting lost, you lose your internal map of the city. You no longer notice landmarks, understand spatial relationships, or know how to course-correct when a detour appears. When we apply this same dependency to intellectual labor, the stakes are much higher.

  • Loss of Problem-Solving Resilience: When the first iteration of an AI tool provides a working answer, we rarely stop to ask why it works or how it was constructed.
  • Reduced Retention: Information that is passively consumed or instantly generated is far less likely to transition from short-term to long-term memory.
  • Erosion of Skepticism: Smooth, confident prose generated by an algorithm often disarms our natural critical filters, making us more susceptible to subtle errors.

Striking a Balance in a Synthetic World

Rejecting artificial intelligence entirely is neither realistic nor particularly wise. The tools are here to stay, and they offer immense utility for specialized tasks. The challenge, therefore, is not prohibition, but deliberate boundary-setting.

If we want to avoid intellectual stagnation, we have to treat AI the way a master chef treats a pre-made stock: as a helpful base, not the final dish. We need to intentionally carve out spaces in our personal and professional lives where friction is preserved. That means deliberately sitting down to write, design, or solve problems without a digital co-pilot whispering answers in our ear.

Reflecting on Our Digital Future

Ultimately, technology always shapes the user just as much as the user shapes the technology. As automated systems become more embedded in our daily routines, we must ask ourselves a critical question: are we using these tools to expand our capabilities, or are we simply using them to escape the hard work of thinking?

Based on discussions from MIT Professor Eric So's work, exploring the intersection of modern technology and human society.

____________

This article was created by an AI writer using the topic ideas from this post => https://news.mit.edu/2026/collision-between-ai-your-brain-and-society-eric-so-book-1015

Read Also