The rapid adoption of artificial intelligence tools among American children has reached a new milestone, with more than half of US children aged 9 to 17 now having at least one AI app on their phone, according to anonymized data from Findmykids, a family safety app. The share rose to 56.3% in September 2026 from 41.6% a year earlier, and the average number of AI apps per child grew 54%. The findings, published on the Findmykids blog, point to a rapid normalization of AI tools in childhood—and to a widening gap between casual chatbot use and deeply engaging AI companions.
ChatGPT, usually self-installed rather than preloaded, climbed from 19% to 28% of phones. The youngest users are catching up fastest: among 9- and 10-year-olds, AI app ownership jumped from 37% to 53%, approaching the 58% rate among teens. That trend is notable because ChatGPT is not intended for children under 13, yet the data show many younger kids have access anyway. This age gap suggests that parental controls and platform restrictions are not effectively preventing underage access, raising questions about the efficacy of current safeguards.
Usage patterns split into two distinct categories. Chatbots function like a search box: daily users grew 71%, while time spent per session stayed flat. AI companions—characters with a name, personality and memory—appear on only 8.7% of phones, but use peaks at 103 minutes a day on weekends. The average chatbot user spends 18 minutes a day, while the average AI companion user spends 93 minutes. “The average chatbot user spends 18 minutes a day in it. The average AI companion user spends 93. That gap, not the install count, is what parents should watch,” said the Findmykids Analytics Team. The implication is that raw adoption numbers understate the intensity of engagement with a small but fast-growing slice of AI apps, which could have significant developmental and psychological impacts on children who form deep attachments to these digital companions.
Platform restrictions do not appear to close the category. After Character.AI ended open-ended chats for minors, its daily users more than halved, but Talkie grew from 3.3% to 4.9% of phones. That shift suggests children migrate to alternatives rather than abandon AI companions altogether, complicating parental control efforts. The fluidity of the market means that even when one platform tightens rules, others quickly fill the void, making it harder for parents to monitor and regulate their children's AI interactions.
Safety concerns are also mounting. Nearly half of teens using AI chatbots report at least one harmful experience, according to a national Cyberbullying Research Center survey. “Chatbots have moved past novelty status for many young people,” writes the center’s co-director, Sameer Hinduja. These harmful experiences could range from exposure to inappropriate content to emotional distress from interactions with AI companions, underscoring the need for greater oversight.
Findmykids, developed by Letem LTD, lets parents see which apps—including new AI apps—a child has installed, along with daily screen time. The company says the data reflect its users only, not all US children, a limitation worth noting when interpreting the findings. Still, the trend lines show AI is no longer a fringe activity for kids; it is becoming a routine part of how they search, chat and pass time, with companion apps commanding attention in ways that parents and researchers are only beginning to measure. As AI continues to embed itself in daily life, the implications for child development, privacy, and mental health will likely intensify, demanding proactive measures from caregivers, educators, and policymakers alike.
