Biking down a country road, behind me was a man on a hissing e-bike, riding fast and heading somewhere important. I was thinking at that speed—25 miles per hour, I estimate—and as dusk fell, he was in danger of colliding with a bear, which would immediately eat him and the bike. But that was not the case.
Later I saw him at the top of the hill I had climbed. He ran around in circles and headed back down while I was resting. The hill was a tough challenge for me. How do you weaken from dependence on a car?I muttered to myself, I'm getting strongeruh. In the world of cycling there is one certainty: Trad bikers will outlive bikers who are fools, giving up the physical benefits, spiritual joys, freedom and independence of human powered machinery.
Our dependence on machines is, of course, growing exponentially as AI becomes widespread. If the Internet, according to the author Nicholas Carrmade us stupid, AI promises to make us even stupider. Carr rightly argued that with its endless distractions and fragmented structure, its blinking rabbit holes, its emphasis on speed and constant switching (between items, links, pages, images, etc.) the Internet causes cognitive damage, rewiring the brain so that we are less able to reflect and meditate, think long and complexly – go deeper. His 2010 book, The Shallows» remains the most important exploration of technological immanence and its consequences since Neil Postman's Technopoly.» (1992).
Now comes AI, and we face a new catastrophe for human cognition as the prospect of AI-induced dementia is confirmed in study after study using large language models such as ChatGPT.
One researcher looked into the “future of critical thinking” in an LLM-saturated environment and found “a significant negative correlation between frequent use of AI tools and critical thinking abilities.” The destruction of critical thinking has occurred due to the apparent increase in “cognitive offload” provided by applications such as ChatGPT. Instead of staying in shape by doing hard work, the “muscles” of the brain atrophy, allowing the machine to bear the load. The study's conclusion had echoes of Carr, noting that dependence on AI could “reduce users' engagement in deep, reflective thought processes.” It found that young people were particularly vulnerable, showing “lower critical thinking scores compared to older participants.” In other words, children, adolescents and young adults are most at risk from these technologies. (Think about it, you adults who make money by selling and promoting technology – you are hurting children.)
Team from Microsoft and Carnegie Mellon University published similar findings in February. The use of “generative AI”, that is, large language models, can “suppress critical engagement in work,” reduce the skill set “for independent problem solving,” and—this should be so obvious that it goes without saying—lead to “long-term over-reliance on this tool.” In this sense, AI is a drug that corrupts the brain; the heavier the use, the stronger the addiction, the greater the damage done.
Finally, in June, a study from the Massachusetts Institute of Technology was published entitled “Your brain in ChatGPT” Study participants who were assigned to write essays were divided into three groups: “brain only,” those who completed the essay without machine assistance; search engine users who were allowed access to Google or the like to supplement the text; and large language model users who were free to use ChatGPT to do the heavy lifting. While writing the essay, the researchers measured “brain connectivity” using electroencephalography. Their conclusion: “Brain-only participants showed the strongest and most distributed networks; Search engine users showed moderate engagement; and LLM users showed the weakest association.”
Cognitive performance decreased with each increase in machine use. Once again, the more you depend on technology to shape your thoughts, the lower your mental abilities, the dumber you become. And stupidity persists over time. “Over a four-month period, LLM users consistently fell behind at neural, language, and behavioral levels. These results raise concerns about the long-term educational consequences of LLM use.”
Between February and April, the number of ChatGPT users worldwide increased from 400 million to 800 million (and this is from 50 million in January 2023). In the UK, the proportion of students who have not used ChatGPT or other LLM programs collapsed from 47% last year to 12% this year, only a tenth of students surveyed opposed the machine. Almost by mid-2024 90% of Harvard students used LLM for study. More than 70% of American adults report regularly using AIand a third says he uses it every day.
Our cognitive catastrophe unfolds as if it were designed in the world, like the sun rising over the horizon and the light enveloping the earth – inevitably, irreversibly, such is the nature of things. This may be so, given that the worship of the god of technology is inherent in industrial civilization and is an expression of the fanatical worship of innovation that is unquestioned in our society.
Meanwhile, I am reminded that the e-bike rider passing us at his artificial speed is turning into a flabby mess by the day. Those few who retain their strength and independence, who refuse to bend the knee to the machine, will only become more powerful and intelligent – perhaps to win the Darwinian race in the long term, when the feeble-minded and slavish dependents of the AI are eliminated after the machine system fails.
Christopher Ketcham is writing a book about environmental rebellion against industrialism. He recently wrote the book This Land: How Cowboys, Capitalism, and Corruption Are Destroying the American West.
