Wellness

We’re offloading mental tasks to AI. It could be making us stupid

Queen Van Bell, a test automation engineer, was symbolizing a living, used artificial intelligence A large Copilot language model for about six months when the Internet fell one day. He was forced to return to his traditional action using him memory He had no experiences, he struggled to remember some of the sentences that coded.

“I couldn’t remember how to work,” said Van Bell, who runs a computer programming company in Belgium, for a salon in a video call. “It has become a method of artificial intelligence … so I had to stop it and re -learn some skills.”

As a director of his company, Van Belle is honored to do a handful of trainees every year. He said that their company had restrictions on the use of artificial intelligence, and the trainees had to limit their use as well. After that, the amount and quality of their coding were dramatically reduced.

“They are able to explain Chatgpt “What they want, generates something and hopes to work. When they enter the real world and they have to build a new project, they will fail,” said Van Bell.

Since artificial intelligence models such as COPILOT and Chatgpt via the Internet in 2022 came, it exploded in popularity, with one reconnaissance In January, it was estimated that more than half of the Americans used Copilot, ChatGPT, Gemini or Claud. Research that examines how these programs affect users is limited because they are very new, but some early studies indicate that they already affect our brains.

“To some extent, these models such as facades or cultivation of brain control – they are strong,” said Kanaka Rajan, a computer nerve and a founding faculty member at the Kimbner Institute to study natural and artificial intelligence at Harvard University. “In some sense, they change the input flows to the networks that live in our brains.”

In February Ticket It was conducted by researchers from Microsoft and Carnegie Mellon University, where groups of people working with data have worked more efficiently with the use of obstetric intelligence tools such as Chatgpt – but they used cash thinking less than the comparison group of workers who did not use these tools. In fact, the more workers tell the confidence of the ability of artificial intelligence to perform tasks for them, the more their critical thinking is reduced.

Last 2024 Ticket It stated last year that the decrease in critical thinking stems from dependence on artificial intelligence to perform a greater percentage of the work of the brain needed to perform tasks in a process called cognitive discharge.

“To some extent, these models are similar to brain control fronts or transplants – they are strong.”

Cognitive discharge is something we do every day when we write our shopping menu, or make an event in the calendar or use a calculator. To reduce the burden of our brain, we can “empty” some of its technology tasks, which can help us perform more complex tasks. However, it has also been linked to the other research For things like a worse memory.

your review It was published in March, “Although laboratory studies have proven that cognitive discharge has benefits to perform the task, it is not without costs.” It is useful, for example, to be able to rely on your mind to remember the grocery list in the event of its loss. So how well the cognitive emptying is for us – and how artificial intelligence accelerates these costs?

This concept is not new: The Greek philosopher Socrates was afraid that The invention of writing will make human beings stupid Because we will not practice our memory as much. It is famous Nothing was written downAlthough his student, Plato, did. Some argue that Socrates was right and the direction is rising: with all the main technological progress, we are increasingly dependent on tools outside ourselves to perform the tasks that we have accomplished at home. Many people may not make routine accounts in their head anymore due to the invention of the calculator, and most people use the GPS (GPS) instead of withdrawing a material map or leaving material marks to guide them to their destination.


Do you want more health and science stories in your inbox? Subscribe to the weekly newsletter of the salon Laboratory notes.


There is no doubt that these inventions made us more efficient, but anxiety lies in what happens when we stop bending the parts of the brain responsible for these tasks. Over time, some argue that we may lose these capabilities. There is an old spirit to “use or lose it” It may apply to cognitive tasks as well.

Despite fears that calculators will destroy our ability to do mathematics, research has generally shown that there Smile difference in performance When using calculators and when they are not. Matthew Fischer, a researcher at the University of South Methost, said that the school system was still in general, spending a lot of time to teach students foundational techniques such as learning multiplication schedules when they can now solve these types of problems with one button.

He explained that this part of the curriculum is an important matter because it provides the founding mathematical building blocks of sports blocks that students learn other parts of mathematics and science. Fischer also told a phone in a phone interview: “If we completely get rid of this sporting basis, then our intuition for subsequent mathematical study, as well as to live in the world and understand basic relationships, will be suspended.”

Other studies indicate that relying on newer forms of technology affects our brain activity. For example, the research found that students The brains were more active when the handwriting information was Instead of writing it on a keyboard and when using Pen and paper versus pen and tablet.

Search also He appears “Using or losing it” is somewhat correct in the context of the skills we learn. New neurons are produced in the hippocampus, which is part of the brain responsible for learning. However, most of these new cells will die unless the brain exerts effort and focus on learning over a period of time. People can definitely learn from artificial intelligence, but the danger lies in abandoning the learning process to simply prepare the information that we feed.

In 2008, about two decades after the general Internet, The Atlantic Ocean published a cover story “Do Google make us stupid?” She asks. Since then, with the emergence of smartphones and social media, research has shown that much time on the Internet can Reducing our ability to focusMake us You feel isolated and Reducing our self -esteem.

One 2011 review I found that people are increasingly resorting to the Internet to get difficult questions and less able to remember the information they found on the Internet when using them to answer these questions. Instead, the participants had a reinforced ability to remember as they found it.

The authors concluded that “the Internet has become an essential form of external memory or whispers, as information is stored collectively outside ourselves.”

People can definitely learn from artificial intelligence, but the danger lies in abandoning the learning process to simply prepare the information that we feed.

In 2021, Fisher participated in his authorship research This was also found that people who used online searches had an enlarged feeling of their own knowledge, as they reported exaggerated claims about the things they read on the Internet compared to the control group that has learned things without them. Description of this phenomenon “Google’s effect”.

Fischer said: “What seems that we are having difficulty doing is the distinction between the place where our elaborate knowledge stops internally and since the knowledge that we can search for only, but we feel as if our knowledge begins,” Fischer said.

Many argue about that This artificial intelligence takes this further It cuts an important part of our fictional process. In opinion piece For internal higher education, John Warner wrote that the excessive of the written tasks was “the risks that hinder the important exploration course of an idea that happens when we write.”

“This is especially true in school contexts, when learning that occurs within the student is much more important than the final product they produce on a specific mission,” Warner wrote.

A large part of the energy intended to understand how artificial intelligence has focused on our brains on adolescents because young generations use these tools more and may be more likely to changes that their brains are still developing. One study 2023, for example, I found young high school students who used artificial intelligence more Less than the ability to adapt to new social situations.

Last 2023 paper It was also found that students who relied heavily on artificial intelligence to answer multiple choice questions summarized reading excerpts less than those who relied on their memory alone, the author of the study Qirui Ju, a researcher at Duke University.

“Writing things helps you really understand the material,” Joe told the salon in an interview over the phone. “But if you replace this process with AI, even if you write high -quality things with the least typographical messages and more coherent sentences, it replaces the learning process so that the quality of learning is lower.”

To get a better idea of ​​what happens with people’s brains when using large language models, Researchers at the Massachusetts Institute of Technology Electrical programming 32 channels for three groups of college age students who were all answering the same writing claims: one group has used Chatgpt, Google, the other used, and the third group simply used their own brains.

Although the study was small, with only 55 participants, its results indicate that large language models can affect our memory, interest and creativity, said Natalia Kosmina, the leader of the “Your Mind on LLM” project, and a research scientist at the MIT Media Lab Lab.

After writing the article, he can remember 85 % of the group that uses Google and the group whose brains use a quote from their writings, compared to only 20 % of those who used large language models, Kos’Na said. Moreover, 16 % of people who use artificial intelligence said they did not even recognize their article as their own after completing it, compared to 0 % of students in the other group.

In general, there was a brain activity and interconnection in the group that used Chatgpt compared to groups that only used Google. Specifically, the activity in the corresponding brain areas of language, imagination and creative writing has been reduced for students who use large language models compared to students in other groups, KOS’NA said.

The research team also conducted another analysis in which students first used their brains before turning into the same task with large language models, and vice versa.

Those who used their brains first and then continued to try their hands in the task with the help of artificial intelligence, and it seems that the performance is better and stimulate the areas mentioned above than their brains. But the same thing was not true for the group that used Amnesty International first and then continued to try it with their brains only.

“It seems that large language models have not necessarily help you and provide any additional delivery in the brain.” “However, there is a possibility … if you already use your mind and then reformulate the task when exposed to the tool, it may be useful.”

Whether artificial intelligence hinders or enhances our learning ability may depend more on how we use it more than if we use it. In other words, Amnesty International is not the problem, but excessive dependence on it.

Van Belle, in Belgium, uses large language models now to write social media posts for his company because he does not feel that this is where his skills are more accurate and the process can take a long time.

He said: “I would like to think I will be able to create a somewhat decent post, but it will take an additional time.” “This is the time that I do not want to waste something that I don’t really care about.”

These days, Amnesty International sees a tool, which can be – as long as we do not empty much of our brain.

Fischer said: “We have been in this fixed march now thousands of years ago, and it seems that we are in the culmination of determining what is left for us to know and do.” “It raises real questions about the best way to balance technology and get the most of them without sacrificing these human things mainly.”

Read more

About artificial intelligence

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button