An alarming new survey from media and tech non-profit Common Sense Media has found that children begin interacting with AI chatbots at an alarmingly early age — with nearly half of 9-12-year-olds having bypassed the 13+ age restriction common to most LLMs.
The survey is titled The 2026 Common Sense Media Census: Tech and Media Use by Tweens and Teens. In the paper, the organization conducted interviews with more than 1,200 American children aged 9 to 17, asking questions about AI use as well as television, gaming, and social media habits. More importantly, the interviews asked these young children how they feel about their own media consumption habits.
Standout findings include the sharp rise in smartphone ownership, with more than half of 11-year-olds already owning a smartphone and near-universal smartphone access by age 17, as well as the incredible amount of time spent on screens: eight hours per day for teenagers and five hours per day for tweens.
Because social scientists were caught unprepared by the massive impact social media has had on teen mental health, there’s a strong push for more studies into the effects of AI interaction on teenagers, especially in the areas of school preparedness and mental health.
A second study, titled “Exploring Artificial Intelligence [AI]: Understanding Relationships between AI Use and Mental Health Among Middle and High School Students,” was commissioned by the JEF Foundation, a mental health advocacy group. They interviewed more than 5,000 American middle and high school students from 21 different schools between Oct. 2025 and April 2026, which resulted in the following four findings, taken verbatim from their report:
-
Students use AI for a wide range of cognitive and emotional support, including turning to AI when they’re struggling emotionally.
-
Certain uses of AI are associated with worse mental health and well-being outcomes, and greater lifetime history of suicidal thoughts and attempts.
-
Students who use AI while emotionally struggling have worse mental health and well-being outcomes, and greater history of suicidal thoughts and attempts.
-
Patterns of AI use differ across demographics, with important implications for mental health disparities.
JED Foundation psychologist Kayla Anderson summarized the findings to NPR: “When young people are using AI, when they’re in emotional distress, they’re more likely to report poor mental health and well-being outcomes,” but she was also quick to stress that the study did not establish causality. In other words, they don’t know if AI is making teen mental health worse, or if teens with worse mental health are gravitating to AI chatbots for help.
Thankfully, there are already a slew of new studies set to conclude soon investigating whether causality can be established. If so, expect lawmakers to come down hard on AI companies to impose more stringent age restrictions and accept greater liability for poor mental health outcomes among teenagers.
