A new study by researchers at the Massachusetts Institute of Technology (MIT) reveals alarming findings about the impact of ChatGPT on human critical thinking.

The article, which has not yet been peer-reviewed, states that the use of large language models (LLMs) can harm learning, especially among young people.

As part of the study, 54 people between the ages of 18 and 39 were divided into three groups. They were asked to write essays modeled on the SAT exam. The first group was allowed to use ChatGPT, the second was allowed to use Google search, and the third was not allowed to use anything. Their brain activity was then recorded using EEG (electroencephalography).

The study found that the participants who used ChatGPT had the lowest brain activity of the three groups. The essays they wrote were similar and lacked original ideas. They showed weaker results at the linguistic, behavioral, and neurological levels. They also became lazier with each subsequent essay. Furthermore, when asked to rewrite their essays without using ChatGPT, they remembered little of their own essays.

In contrast, the group that used neither ChatGPT nor Google Search showed the highest neural connectivity. Participants in this group were more curious and engaged, and also expressed greater satisfaction with the essays they wrote.

Although the study has a small sample size and has not yet been peer-reviewed, it is essential to investigate the impact of this technology on the brain at this early stage.

An earlier study also found that the more a person engages with ChatGPT, the lonelier they become. Another study from March showed that although generative artificial intelligence increases people's productivity, they also become less motivated.

These observations come at a critical time, as the use of artificial intelligence for various activities is growing, and it is crucial to study its impact on the brain. | BGNES