Two recent polls show that a significant minority of young people are turning to AI chatbots to discuss their everyday fears and concerns.
A study published in JAMA network openis based on responses from 1,058 U.S. adolescents and young adults ages 12 to 21 and found that 13.1 percent reported using generative AI chatbots for mental health advice. The Pew Research Center published similar findings in 2025, reporting that 12 percent of teens sought emotional support or advice from an AI chatbot.
Proponents of the technology argue that AI provides a space for self-reflection, venting frustrations, and seeking modest reassurance.
Digital safety experts say this shift is indicative of broader societal trends, rather than simple curiosity about new technology. It also suggests a fault line in generational attitudes around AI.
Parents may be concerned if they notice that their children are having emotional interactions with AI. They may instinctively restrict access or react suddenly. Experts advise against sudden negative reactions.
Dr Melissa Morrison, an educational and child psychologist based in the UK’s West Midlands, says a parent’s initial response can make a difference in whether a child feels safe enough to explain.
“If parents move in too harshly and critically, children stop talking,” Morrison says.
She explains that a calm approach using open questions can help parents determine whether their child is seeking reassurance, information, or a personal outlet.
Psychologist Amena Zaheer says young people can be reluctant to discuss vulnerability with their parents for fear that they will overreact.
AI may seem like an easier option because it removes the pressure of a human response. Acknowledging this dynamic does not mean encouraging reliance on technology, Zaheer said. Instead, it means recognizing that the emotions that drive the interaction are real.
Experts also warn that confident language used by AI can be mistaken for reliable advice.
While chatbots often communicate clearly and authoritatively, experts stress that AI does not interpret emotional nuances or personal context the way humans do.
Brazil-based AI ethicist Catalina Doria says this distinction is important when children seek guidance.
“AI systems generate language, not judgment,” she says.
Answering fluently and without hesitation gives the impression that you understand. Doria explains that AI typically generates responses by predicting possible word patterns, rather than assessing what is appropriate or safe for a particular individual.
“This system does not assess vulnerabilities or risks,” she added.
Encouraging children to question answers, verify information, and discuss sensitive advice with trusted adults is an important safeguard.
“Kids need to understand that AI doesn’t ‘know’ them,” she says.
Psychological safety is just one issue. Digital privacy is another concern.
Cybersecurity experts have warned that many young users misunderstand how private AI conversations are. Although the interactions may feel confidential, AI platforms do not work like discussions with parents, teachers, and clinicians.
Andrei Sidenko, lead web content analyst at cybersecurity firm Kaspersky Lab, said the information shared with the AI system will be processed to generate a response, and elements of those interactions may be saved depending on the platform’s policies.
“In some cases, third-party service providers involved in running or supporting the platform may have access to stored data,” he says.
If our systems are compromised, any personal information held on us is at risk.
“Third party involvement increases exposure in the event of a security breach,” Sideenko says.
Parents should not be afraid to make their children understand these boundaries.
“Parental awareness and involvement is very important,” he says.
Having clear, age-appropriate conversations about what AI is, how it works, and what should remain private can turn uncertainty into understanding, he added.
Noor Ibrahim is the 2025-26 Rosalynn Carter Mental Health Journalism Fellow..
