The MIT Media Lab in Cambridge, USA, is a hub of futuristic innovation. With its glass cabinets showcasing prototypes of quirky inventions, from miniature robots to AI-generated sculptures, it's a place where the future feels tangible. But amidst this technological wonderland, a pressing question arises: Are we embracing a golden age of stupidity? As AI assistants like Oscar guide us through waste sorting, and researchers like Nataliya Kosmyna develop brain-computer interfaces to aid communication for the speech-impaired, a deeper concern emerges.
Kosmyna, a trendsetter in a blue shirt dress and a vibrant necklace, has noticed a peculiar trend. People are reporting changes in their brain function after using large language models like ChatGPT. They claim their memories are deteriorating, and Kosmyna's own observations support this. Colleagues and job applicants are increasingly reliant on AI, even during interviews. This prompts a crucial question: Are we outsourcing our thinking to machines?
In an experiment, Kosmyna and her MIT colleagues monitored brain activity while participants wrote essays with and without digital assistance. The results were startling. Those who used ChatGPT exhibited significantly reduced brain connectivity, particularly in areas associated with cognitive processing, attention, and creativity. This raises concerns about the long-term effects of AI on our cognitive abilities.
The implications are far-reaching. Students are using AI to complete homework, leaving teachers worried about their learning. The ease of access to technology is making us increasingly dependent, and our brains, which thrive on challenges, are being deprived of the friction needed for learning. We're outsourcing more and more tasks to our devices, from simple calculations to fact-checking, and this reliance is making us less engaged with the real world.
The concept of a "stupidogenic society" is gaining traction, where machines do our thinking for us, leading to a decline in human intelligence. This is evident in falling test scores and IQ levels in developed countries. As we become more reliant on technology, we're outsourcing not just memory and data processing but also critical thinking and creativity. The convenience of AI is luring us into a state of passive dependence, where we're users first and thinkers second.
However, the relationship between technology and intelligence is complex. Writing, for instance, has transformed how we access and retain information, enabling us to achieve more complex tasks. But the internet, with its endless distractions, often leads to cognitive overload. We're constantly bombarded with notifications, and our attention is monetized. This creates a state of continuous partial attention, where we're always on edge, making us forgetful and less attentive.
Generative AI has introduced a new dimension to this dilemma. It allows us to outsource thinking itself, not just memory and data processing. As we transition from the internet era to the AI era, we're consuming low-value, ultra-processed information that bypasses essential human functions like assessing and summarizing. This raises concerns about the quality of our thinking and the originality of our ideas.
The impact of AI on critical thinking is evident in classrooms, where students are using AI to complete assignments, leading to a decline in the quality of discussions. The anchoring effect, where AI answers set our brains on a specific path, limits our ability to consider alternative approaches. This is particularly concerning in education, where critical thinking and problem-solving skills are essential.
The fear is that we're creating a generation of mindless, gullible AI essay-writing drones, rather than creative, original thinkers. The education system, once a bastion of knowledge and understanding, is at risk of becoming a factory for producing AI-dependent graduates. The question remains: Are we sacrificing our intellectual independence for the convenience of AI, and if so, at what cost?
