2024-02-14 22:21:00
In his article titled “Catastrophic Effects,” Professor Umberto León Domínguez of the University of Monterrey explores the potential implications of artificial intelligence (AI) on human cognitive functions. Looking at chatbots such as ChatGPT, designed to mimic human conversation, Domínguez expresses concerns about the possibility of AI acting as a “cognitive prosthesis,” performing cognitive tasks for humans.
He warns against “cognitive offloading,” warning that over-reliance on AI could lead to a weakening of human cognitive abilities, compromising critical thinking and creativity. Despite these advances, AI will never be able to surpass the capabilities of the human brain. Researchers at the University of Sheffield have pointed out that there are inherent differences that prevent AI from thinking like humans.
Furthermore, two crucial distinctions were highlighted. First, the human brain is intrinsically linked to physical elements, allowing direct perception and interaction with the real world. This distinction plays an important role in differentiating AI from the human brain. Furthermore, our brains are developed according to a specific architecture, which is not the case with AI models. According to the study, AI can learn complex patterns from data, but it lacks the connection with the real world, which plays a major role in terms of thinking.
Although some see cognitive offloading as beneficial, freeing up cognitive abilities for more complex tasks, Domínguez highlights the risk that technologies like ChatGPT will make individuals complacent and neglect essential cognitive tasks. In comparison to other technologies, ChatGPT is considered a significant amplifier of cognitive offload due to its ability to autonomously generate ideas and decisions. Domínguez calls for awareness of the need to actively stimulate intellectual abilities, emphasizing that cognitive development requires active participation and cannot rely solely on technological assistance.
Sent by Umberto León Domínguez
The Retrained Generative Transformer (ChatGPT) represents a revolutionary advancement in artificial intelligence (AI-chatbot) technology, using transformation algorithms to improve natural language processing and making it easier to use them to address specific tasks. These AI chatbots can answer questions by generating verbal instructions similar to those a person would provide during the process of solving a problem.
Objective: ChatGPT has become the fastest growing software in history in terms of user adoption, suggesting widespread use of this technology among the general population. Current literature primarily focuses on the functional aspects of these technologies, but the field has yet to explore hypotheses about how these AI chatbots might impact evolutionary aspects of human cognitive development.
Thesis: The “neural recycling” hypothesis posits that the brain undergoes a structural transformation by incorporating new cultural tools into “neural niches”, which modifies individual cognition. In the case of technological tools, it has been established that they reduce the cognitive demand necessary for solving tasks through a process called “cognitive offloading”. In this theoretical article, three hypotheses were proposed through prospective inference on how algorithms such as ChatGPT and similar models may influence the cognitive processes and structures of future generations.
Conclusions: By predicting the neurocognitive effects of these technologies, the educational and policy communities can anticipate future scenarios and formulate strategic plans to mitigate or enhance the cognitive influence that these factors may have on the general population.
Domínguez’s interest in ChatGPT stems from its potential as a technological milestone. He sees it as a sign of the technological singularity, a concept that suggests that the development of AI could reach a point where it begins to advance beyond human control, potentially merging human and machine intelligence. .
“As a university professor, I see my activities as intellectual challenges aimed at stimulating and training cognitive functions useful in the daily lives of my students, such as problem solving and planning skills,” explained Domínguez, who is also director of the Human Cognition and Brain Studies Lab and researcher at the Artificial Intelligence Group.
“The emergence of a tool like ChatGPT made me concerned that it might be used by students to accomplish tasks, thereby preventing the stimulation of these cognitive functions. From this observation, I began to explore and generalize the impact, not only as a student but also as humanity, of the catastrophic effects that these technologies could have on a significant part of the population in blocking the development of these cognitive functions. »
The frontiers of AI: cognitive prosthesis or threat to human cognitive independence
One of the paper’s most striking claims is that AI can act as a “cognitive prosthesis,” a concept introduced in a 2019 study by Falk Lieder and colleagues. In essence, this means that AI could perform cognitive tasks on behalf of humans, much like a prosthesis is used to replace a lost limb. It’s not just about simple tasks like calculating numbers or organizing schedules. Research suggests that AI capabilities could extend to more complex cognitive functions, such as problem solving and decision-making, traditionally considered hallmark traits of being human.
When asked about machine learning and its contributions to cognitive science by the Massachusetts Institute of Technology (MIT), Noam Chomsky and Steven Pinker, two of the world’s leading linguistic and cognitive scientists, expressed much skepticism and some disappointment. “In almost all relevant fields, it is difficult to see how machine learning can make any contribution to science,” laments Chomsky, “particularly to cognitive science, whatever value it may have in constructing useful devices or to explore the properties of the computer processes employed. »
Lieder and his colleagues highlighted scenarios in which people’s natural penchant for short-term rewards steers them away from actions that would be more beneficial in the long term. For example, choosing to watch TV and relax instead of working on a difficult but rewarding project. To address this problem, they proposed using AI to “gamify” the decision-making process. Gamification is the process of adding game-like elements, such as points, levels, and badges, to non-game activities.
Through a series of experiments, Lieder and his colleagues provided the first evidence of the benefits of this approach. They found that AI-enhanced decisions helped individuals make better choices faster, reduce procrastination, and focus more on important tasks.
However, Domínguez’s article warns of the potential risks of integrating AI into our cognitive processes. One of the main concerns is “cognitive offload,” where humans could become overly dependent on AI, leading to a decline in our ability to perform cognitive tasks independently. Just as muscles can weaken without exercise, cognitive abilities can deteriorate if not used regularly.
Opinions differ on this subject. Some express concern about a possible decline in human intellectual abilities due to an increasing reliance on AI, highlighting the risk of “cognitive offload” and the negative consequences on critical thinking and creativity. On the other hand, some take a cynical approach, suggesting that humanity is already losing intelligence.
By recognizing AI as a catch-up opportunity for those who have yet to assimilate certain concepts, they humorously approach the idea that future AI masters could be beneficial to society. Reference to ancient thoughts, such as that attributed to Socrates, raises ongoing concerns about how new technologies may influence the way we think and remember information.
Source : “Catastrophic effects”, Professor Umberto León Domínguez of the University of Monterrey
And you ?
What is your opinion on the subject?
Do you think chatbots, such as ChatGPT, can actually act as “cognitive prosthetics” and perform cognitive tasks for individuals?
See as well :
The limits of artificial intelligence: why AI can’t outperform the human brain, according to University of Sheffield researchers
MIT researchers just discovered AI that mimics the brain on its own, machine learning is starting to look a lot like human cognition
How will we know that we have achieved a form of artificial general intelligence? For some AI experts, this will be when the machine has human-level cognition
1707968377
#technology #lead #decline #intelligence #Researcher #Umberto #Domínguez #University #Monterrey #expresses #fears