A study by researchers at the Massachusetts Institute of Technology (MIT) indicates that reliance on AI chatbots, such as ChatGPT, may impair the development of critical thinking, memory, and language skills.

This research raises concerns about the long-term effects of AI tools on education and workplace performance.

The study involved 54 participants divided into three groups, each tasked with writing essays using different methods: ChatGPT, internet search engines, or their own knowledge.

Brain activity was monitored through electroencephalogram (EEG) scans. The results showed that participants who relied on ChatGPT exhibited reduced brain connectivity and lower theta brainwaves associated with learning and memory.

Participants using chatbots struggled to recall essay content. An alarming 83% were unable to provide accurate quotes from their work, compared to only 10% in the non-AI groups.

The study suggested that frequent use of AI tools may lead to skill atrophy, diminishing abilities in brainstorming and problem-solving. These findings contribute to the ongoing debate about AI’s role in education and its effect on cognitive engagement.

OpenAI, the developer of ChatGPT, has not commented on the study.

.source-ref{font-size:0.85em;color:#666;display:block;margin-top:1em;}a.ask-tia-citation-link:hover{color:#11628d !important;background:#e9f6f5 !important;border-color:#11628d !important;text-decoration:none !important;}@media only screen and (min-width:768px){a.ask-tia-citation-link{font-size:11px !important;}}

🔗 Source: The Telegraph

🧠 Food for thought

1️⃣ Educational technology has historically shown mixed impacts based on implementation

The MIT study on AI reflects a broader pattern in educational technology where implementation matters more than the technology itself.

Previous research analyzing 126 studies found that simply providing computers and internet access doesn’t significantly improve academic outcomes, despite education technology spending exceeding $13 billion in the US alone 1.

Educational software has shown promise (with 20 of 30 studies reporting positive effects), but only when thoughtfully integrated with clear learning objectives 1.

This historical context suggests that the cognitive concerns about AI chatbots may similarly depend on how they’re implemented rather than representing an inherent harm of the technology itself.

The research consistently shows that technology is most effective when used as a supplement to good teaching rather than a replacement, with collaborative learning approaches showing particular promise 2.

2️⃣ Moderate technology use correlates with better outcomes than high usage

The MIT findings on reduced brain activity align with previous research indicating an optimal “middle ground” for technology use in learning.

A study analyzing PISA and NAEP data found that fourth-grade students using tablets in all classes scored 14 points lower on reading exams than those who never used tablets, suggesting excessive technology use may be counterproductive 3.

This data revealed that moderate technology use was consistently associated with better performance compared to high usage across multiple subjects and grade levels 3.

The brain connectivity findings in the MIT study (showing reduced theta waves associated with learning) provide a potential neurological explanation for this phenomenon, where some technological assistance enhances learning without creating dependency 4.

These patterns suggest that educational institutions may need to establish guidelines for appropriate AI usage frequency rather than focusing solely on whether the technology should be used at all.

3️⃣ Global AI education adoption is outpacing our understanding of its impacts

The concerns raised in the MIT study come amid rapid global implementation of AI in education, with 86% of students already using AI tools in their studies despite limited research on long-term effects 5.

This widespread adoption is happening despite significant uncertainty among users – surveys show 53% of students worry about incorrect information from AI, and only 7% of adolescents report using AI daily, suggesting a knowledge gap in effective utilization 5, 6.

Nations are making major policy commitments despite these uncertainties – Estonia plans to integrate ChatGPT Edu into schools nationwide by September 2025, and Beijing is introducing AI courses for primary students as part of national AI development efforts 7.

The disconnect between rapid adoption and limited understanding is particularly concerning given that 71% of teachers report never having used AI tools with students, creating potential misalignment between student practices and instructor guidance 8.

Without proper training and guidelines, we risk implementing technologies that may compromise the critical thinking and memory skills highlighted in the MIT research before we fully understand their impacts.

……