The Conversation: half of children use AI

The Conversation reported on August 5 that 54% of Australian children ages 10 to 17 are using AI tools for personal and social reasons, citing new data from the country’s eSafety Commissioner. The article said that use goes beyond homework and information searches into advice about health, feelings, relationships and everyday decisions. The survey adds to a broader run of 2026 research showing that young people are not only using chatbots widely, but are also turning to them for support that once might have gone to friends, parents, teachers or clinicians. ### How large is the shift from practical AI use to personal use? Australia’s eSafety Commissioner surveyed 1,950 people ages 10 to 17 in February and March 2026, and 78% said they had used an AI assistant such as ChatGPT, Gemini or Claude, according to The Conversation. Within that group, 54% said they had used AI for personal or social reasons, and about 20% said they used AI tools at least daily. (theconversation.com) The Conversation said the practical-to-personal drift matters because many children first approach AI for schoolwork, information or entertainment, then begin using the same systems for more intimate conversations. That pattern is consistent with U.S. findings from Pew Research Center, which reported in February that a majority of U.S. teens use AI chatbots and about three in ten do so daily. (theconversation.com) ### What are children and teens actually asking AI about? The eSafety data cited by The Conversation showed that 33% of those using AI for personal reasons asked for advice about physical health, 33% asked what to do in a particular situation, 22% chatted about feelings or challenges, and 20% asked for advice about mental health and wellbeing. Another 17% said they used AI to hang out or chat, 12% for role play, and 7% of teens ages 13 to 17 for flirting, romance or dating. (theconversation.com) Around 10% of teens also used AI companions, The Conversation said, referring to tools designed for sustained or relationship-style interaction. The article said children with disability and LGBTQIA+ young people were more likely than others to engage with AI companions. (theconversation.com) ### Is this only an Australian story? Pew Research Center reported on February 24 that just over half of U.S. teens said they had used chatbots for help with schoolwork, and 12% said they had gotten emotional support from the tools. Pew surveyed 1,458 U.S. teens and their parents from Sept. 25 to Oct. 9, 2025. RAND said on June 1 that 19.2% of U.S. adolescents and young adults ages 12 to 21 reported using AI chatbots for mental health advice, up from 13.1% a year earlier. (theconversation.com) RAND said nearly two-thirds of those users, 63%, had not told anyone they were using chatbots that way, and 92% described the advice as somewhat or very helpful. (pewresearch.org) Common Sense Media said in its March 2026 “Generation AI” report that families share concerns about dependence and about misuse of personal data. That report focused on U.S. kids ages 12 to 17 and their parents. ### What are the concrete risks in turning to chatbots for support? The Conversation said privacy is one immediate concern because children seeking help about health, feelings or personal problems may disclose sensitive information to AI systems. (rand.org) The article also pointed to the risk that young users may treat AI outputs as trustworthy or confidential when those systems are not substitutes for trained adults or clinical care. (commonsensemedia.org) RAND researchers Ryan K. McBain and Jonathan H. Cantor said the private nature of this use is part of the concern. McBain said AI chatbots are already part of how many young people seek mental health advice, while Cantor said many appear to be doing so without the knowledge of parents, clinicians or other adults. (theconversation.com) ### Why would this matter in forensic or clinical assessment? Digital histories already include texts, search records and social media, and AI chats may increasingly belong in that record when an evaluator is trying to understand a young person’s beliefs, disclosures or distress. That is an inference from the pattern described by The Conversation and RAND: if adolescents are using chatbots for advice about health, emotions and stressful situations, those interactions may become relevant to later evaluations of decision-making, risk, secrecy or reported symptoms. (rand.org) The next public reference points are the underlying eSafety Commissioner material cited by The Conversation and continuing U.S. tracking from groups such as Pew, RAND and Common Sense Media, which have all published 2026 survey data on youth AI use. (theconversation.com)