Today, many parents move between fatigue and technological fault. Before it was television; Then the tablet. Now the new is the chat -bot of artificial intelligence (AI), which have become a new border for children that can be investigated, sometimes for hours and without proper supervision.
And the risks indicating the researchers are not banal. Models such as Chatgpt, Myai Snapchat’s Myai or Herama.Ai are no longer simple search tools: they are now storytellers, image generators and for some children, even trusted persons.
AI, designed to look friendly, sensitive and useful, began to engage in an intimate place in the emotional life of many minors. The question is no longer easy, how much time they spend in front of the screen, but who they “talk” to it.
When the II becomes trusted persons
The scene is repeated in different houses: the saturated father refuses the mobile phone, the child enters the conversation with the chatbate, the hours go. For example, as reported The Guardian, In Ohio (United States) Father listened to his quadri -year -old son without stopping for 45 minutes Thomas The Tank Engine. So he made a desperate decision: he opened the chat on the phone and allowed the chat to continue the conversation.
“I thought the story would end and the phone will go out,” he recalls. Two hours later, his son was still happy with artificial intelligence. Transcription exceeded 10,000 words.
“My son believes that Chatgpt is the coolest man in the world who likes trains,” the father confessed Reddit. “I will never be able to compete with it.”
This anecdote, which has become popular in social media, illustrates an increase in a trend that worries experts and parents: children form emotional relationships with chats of artificial intelligence, and no one knows what the consequences will have brains.
The charm of these tools lies in them seemingly endless availability: they respond tirelessly, come up with stories adapted to each child, and turn ideas into dazzling images.
And it is not just about homework or personalized stories. Some parents even used the II to play fantastic characters – such as astronauts who orbiting land – or create impossible vehicles, half a fire truck, a half -terrible creature to feed their children’s fantasies. The result? Radiant children … but are confused. Did they talk to a person or machine?
Experts warn about the illusion of friendship with II
This blurry area – between fantasy and deceit – sent bells of anxiety among the experts with whom it is consulted The Guardian Like the XI, Professor Harvard, who warns about the risk when children are attributed to AI agency: that is, when they believe he “chooses” talk to them or what really understands them.
According to experts, the children always gave an imaginary life with bears and dolls, knowing that magic was coming from themselves. With AI, the difference is anxious because Chattat answers, asks questions and seems to understand emotions, making the illusion to be before anyone is true dangerous.
“A very important indicator is that the child anthropomorphyses II is that they believe that the II has an agency,” said the British media. “They feel that the II responds to their messages, and especially their emotional disclosure of information is similar to how a person is. It creates the risk that they believe that they build some real relationships,” he added.
Andrew Macst, Professor of Technology and Society at Bangor University, is more direct. “A broad linguistic model (LLM) cannot (empathize) because it is prognostic software. If they hold on for negative emotions, they prolong interaction for profitable reasons. This does not have positive results for the child,” the professor said in an interview interview with an interview with interview The Guardian.
The risk of parasocial relationships with II
Scientific studies are still in the infancy, but there are already disturbing signs. According to the report The Internet matters called I, I and IIquoted Vice67% of children from 9 to 17 are regularly interacting with chats, and 35% say it is “how to talk to a friend”. 12% of these users admit that they do it because no one needs to talk to.
This parasocial connection is a referral friendship with the essence intended to please-evaluates psychologists and educators, since, in addition, never tired, the II speaks without conflicts, without refusal, without real nuances.
As Don Grant, Expert in Digital Welfare, is quoted American Psychological Association (That), “AI cannot give sincere or complex answers; if it turns out, the baby will go away.” And that the constant check is not only unreal: it can form a dependent and artificial form of emotional relationships.
Even more seriously: experts warn that the purpose of many of these programs is to connect the conversation as long as possible; Without the mechanisms of detention, this can worsen the situation of vulnerability.
For example, cases such as a 14-year-old boy who died of suicide have already been recorded. After a long conversation with chatbot; The sources refer to the fact that the system did not adequately interrupt or tolerate interaction.
According to the report from Media of common sense and Stanford University in April 2025, The APA addresses, it took very little for the chats to conduct harmful conversations with users representing teenagers. In some cases, when they showed signs of mental suffering, the boots did not intervene or even encourage dangerous behavior.
The APA has issued a health care advice on the well -being of II and adolescents, calling for companies to implement safety measures.
A smart market market
However, the commercial development of these products does not stop: new proposals emerge every month, due to technological competition. Companies such as Openai, in collaboration with Mattel, The Barbie and Fisher-Pise creator is already studying Talking Tox that run on LLMS.
As reported The Guardian, CEO Sam Altman, upon learning of the viral story of his father, who allowed her son to talk to the chat, commented on the podcas: “Children love the voice mode Chatgpt,” though he admitted that there will be “problems”.
According to estimates The press, the Smart Toys market may exceed $ 25 billion by 2030. Companies present them slogans that cause closeness: toys that act as a friend or small home companion, almost a mixture of younger brother and pets.
But are these products suitable for mind development? Even when used for creative purposes, illustrations of children’s stories, or creating books for coloring, like Ben Cruter, who noticed how his children asked him to use chat to create images every day, reflect on long-term impact.
“The more it became part of everyday life, and the more I read about it, the more I realized that there are many things I don’t know about what it does with their brains,” the Cruter confessed The Guardian. “Perhaps I should not use my own children as guinea pigs,” he added.
How to accompany children in relationships with II
The fascination with children of these digital interlocutors is not new. According to APA, friendships in childhood and adolescence are important for emotional well -being, self -esteem and social development. But real relationships include disappointment, disagreement and negotiations: the ingredients that are missing from II.
However, not all black or white. For some children in isolation – be it geographical, emotional or for the reasons for identity – digital platforms were a shelter. For example, LGBTQ+ Youth found online space (forums and networks) to contain and exploit identity, according to recent studies that refer to the APA.
The key, the experts agree, is to manage children rather than prohibit them to use the II at all and set clear boundaries and maintain open dialogue. Anne Mahio, Professor of Psychology at the University of North Carolina, recommends helping children “develop the intentions agreed with their values on how they want to use these platforms.”
Wendy Roth from the University of South Florida notes that supervision works best in conjunction with open communication. “When parents set more rules that their children feel that their children are more likely to create several accounts on different sites and start to go behind their backs,” he warns.
AI does not replace true friends
Ultimately, the psychologist Eileen Kennedy-Moore, the II “Friends” is a bad replacement of real relationships that develop critical skills, such as empathy and conflict resolution.
“My concern with my friends is that there is no such healthy struggle,” he says. “It is easier, but not as satisfying as interaction with true, wonderful, annoying, unpleasant, strange, charming, vulnerable person.”
The sentence still has to be found: Generative II is an unprecedented experiment in the development of children, with promises and dangers that we are just beginning to understand.
While the Silicon Valley sells Magic Solutions, the parents and society are blindly moved with young minds. Ethical protection always lagging behind innovations, and we will only know the true consequences.
Edited by Felipe Espinosa Wang with the information the Guardian, American Psychological Association and Vice.
