## The Erosion of Thought: How AI Builds on a Foundation of Internet-Induced Cognitive decline
A growing body of evidence suggests that reliance on artificial intelligence, specifically large language models (LLMs), is negatively impacting cognitive function. This concern isn’t emerging in a vacuum; it builds upon existing anxieties about the internet’s effect on our ability to think critically and independently.
Concerns about the impact of AI began surfacing early in its rapid adoption. As early as February, warnings arose that the use of “generative AI” could “inhibit critical engagement with work,” diminish skills ”for self-reliant problem-solving,” and foster “long-term overreliance on the tool.” The author frames this potential outcome as a dangerous addiction, where increased use leads to greater dependence and cognitive damage.Scientific research is beginning to support these fears. A study published by MIT in June, titled “Your Brain on ChatGPT,” investigated the neurological effects of AI-assisted writing. Participants tasked with writing an essay were divided into three groups: those writing without assistance, those using search engines, and those utilizing ChatGPT. Researchers measured ”brain connectivity” using electroencephalography and found that “Brain-only participants exhibited the strongest, moast distributed networks; Search Engine users showed moderate engagement; and LLM users displayed the weakest connectivity.”
The study demonstrated a clear correlation: increased reliance on AI corresponded with decreased cognitive activity. Researchers found that this “stupifying” effect persisted over time, with “LLM users consistently underperformed at neural, linguistic, and behavioral levels” even after four months. The results “raise concerns about the long-term educational implications of LLM reliance.”
The speed of AI adoption is accelerating this potential cognitive decline. Between February and April, the number of ChatGPT users worldwide increased from 400 million to 800 million, following an initial jump from 50 million in January 2023. In the U.K., resistance to using these tools among students has plummeted, with the proportion *not* using ChatGPT or similar LLMs collapsing from 47% last year to just 12% this year. at Harvard, nearly 90% of students are now employing LLMs for their studies. This trend extends beyond academia, with over 70% of American adults reporting regular AI use,and a third using it daily.
The author suggests this widespread adoption feels inevitable, a outcome of a society that unquestioningly embraces technological “innovation.” However, a counterpoint is offered: those who resist this dependence, maintaining their cognitive strength and independence, may ultimately be better positioned to thrive when the limitations of AI and the systems supporting it become apparent.
*Christopher Ketcham is writing a book on environmental revolt against industrialism. He is the author, most recently, of “this Land: How Cowboys, Capitalism, and Corruption Are Ruining the American West.”*
