Over-reliance on AI Numbing Students’ Minds

The integration of Artificial Intelligence (AI) into education, research and daily life has attracted significant attention in recent years. By streamlining repetitive tasks, supporting data analysis and introducing new learning methods, generative dialogue systems have become increasingly popular, especially among students and young professionals. With the rapid development of the system, how people use AI has transformed significantly. It is undeniable that the generative system plays a pivotal role in providing important information and analysis to users, especially students and professionals. However, the concerning issue is students’ overreliance on AI. The issue is not limited to global students only; rather, it is a concern for learners in Bangladesh, particularly in the higher education sector. 

Multiple global studies have found that over-reliance on AI can have a serious cognitive impact, as it weakens critical thinking ability, attention span, creativity and reflective thinking skills. According to a study by MIT Media Lab, ChatGPT users have the lowest brain engagement and consistently underperform at neural, linguistic and behavioural levels. The study found that these users often resort to copy-and-paste.

AI platform is doing all the mental labour on behalf of the students, which is forming a barrier in the learning process. Learning can only occur when the brain is actively engaging with what the learner is trying to make sense of and learn. Unfortunately, students using Large Language Models (LLMs) to complete their classwork and assignments are hindering their learning. The alarming fact is that these students are not merely using LLMs for assignments but also for group work, often relying on AI instead of collaborating with peers. This over-dependence may hinder their ability to interpret subtexts and articulate their own opinions, negatively affecting future learning.

The issue is compounded when some educators overlook or encourage such practises, implicitly permitting continued reliance on AI. This may result in future professionals who struggle to produce basic reports without digital assistance. Maisha Rahman (not her real name), a student from a private university, commented, “I have unfortunately witnessed firsthand how many students nowadays cannot imagine writing a single paragraph or speaking three to four lines on a given topic without seeking AI for input, including subject matters that most of us have learned at elementary levels like climate change, history or even just our own country’s food or culture. The hand moves to pull out ChatGPT on the phone before the brain even attempts to understand the prompt.”

Pulling out generative dialogue apps before the brain can fully comprehend a situation is a dangerous habit for both students and the economy. Why? Because it is not normal for people to need AI to converse or write about regular topics. The only time students might need help from LLMs is when they are doing extensive research on difficult topics, but they should not seek generative systems to navigate their lives. AI has already started to numb their minds, and it is plainly wrong. They have forgotten that the mental work, the struggle, the false starts, and the moment something finally clicks are what our brain need. Using AI for your learning is like going to a gym and letting the robot run on the treadmill. You get nothing! 

A recent study found that heavy AI users scored significantly worse on a standard critical thinking test because they are so used to offloading their thoughts to robots. Another earlier study has found that AI can mess with people’s ability to retain information. Those who often use these LLMs suffer from memory loss too.

LLMs, like Claude, ChatGPT and Gemini, are remarkably good at generating ideas, and that is the problem because people often think that they are being more creative with AI, but in reality, it has been the opposite. Research has found that people who use AI for creative tasks tend to produce ideas that are more predictable and less original than people who do not. This is exactly what is happening with these university students. They think the generative system is helping them produce unique and creative assignments and presentations; in reality, it is the opposite. The way these students are using the generative system spells a future full of identical, soulless slop, devoid of passion, effort or human touch. 

The human brain is structurally different from LLMs in ways that really matter. It creates connections that are personal, unexpected and genuinely novel in ways that digital probability machines simply cannot replicate. People, primarily students, love cognitive shortcuts, and that is why AI has become so popular, but continuous challenge is required for long-term brain health. Educators and the government must realise that over-dependency on AI is harming future generations rather than helping them. Educators must discourage AI usage in classrooms. Without adequate policy implementation, the generative dialogue system will continue to erode the young talent pool.