
GenAI use among US youth shows wide variation
Key Takeaways
- Nearly one-third of US youth used a GenAI app, but most engagement was brief, with heavy use concentrated in a small subset of users.
- GenAI use increased with age, peaking among midteens, and commonly occurred after school, though use during school hours was also reported.
A study found that while nearly one-third of US youth used generative artificial intelligence apps, usage patterns varied widely by age, timing, and intensity.
A cross-sectional study published in JAMA Network Open has highlighted significant variation in generative artificial intelligence (GenAI) application (app) use among US youth, with some use reported in approximately half and heavy use reported in a small subset.1
GenAI, a tool capable of generating content, has been reported as the fastest adopted technology among adults.2 However, there is little data about the use of this tool among children and adolescents.1 Additionally, as some adults have reported using GenAI for emotionally intimate conversations, concerns have arisen about socioeconomic reliance from heavy-using youth.
“Understanding patterns of GenAI use among youth—including across different ages and times of day—is a critical first step toward evaluating the developmental risks or opportunities presented by this technology,” wrote investigators.
Study design and data collection
The study was conducted to evaluate usage patterns of GenAI use among youths. The Aura parental monitoring app, which allows parents to connect to internet-capable devices used by their children, was used to obtain relevant data, which was aggregated for each child across all devices.
Keystroke data was evaluated to categorize the usage of all apps, alongside supplemental virtual private network data for ChatGPT use. Data between September 1, 2024, and April 1, 2025, was collected for the analysis, excluding data from holiday breaks.
Individuals with at least 14 total days and at least 10 school days of keyboard data were included in the analysis. The year of birth for these participants was reported to determine their age. Top-down and bottom-up methods were both utilized to determine GenAI and social app use.
Temporal categories included weekday, weekend day, school hours, after school, and nighttime. Nighttime was determined based on typical sleep onset and late-night media use. These temporal analyses were conducted to assess when GenAI use may replace healthy behaviors such as sleeping or paying attention in school.
GenAI app usage
There were 6488 US youth aged 4 to 17 years included in the analysis, 31.9% of whom used a GenAI app from their device and 25.9% had at least 3 minutes of use. Of participants, 1632 accessed ChatGPT, with a rate of 78.8% among GenAI users. This made it the most common app used in the study population.
Apps marketed to provide social companionship were also frequently accessed with these apps comprising 7 of the 17 apps used by 10 or more participants. Gen AI use during school hours was reported in 21.3% of participants and after school use in 25.5%. A slight decline was noted for nighttime use in comparison, with a rate of 12.5%.
GenAI use based on age category was also reported. Rates included:
- 50.4% of midteens
- 42% of young teens
- 20% of preteens
- 5.6% of young children
Duration and intensity of use
An average 2.37 minutes per day of GenAI use was reported among users. However, the median was 0.18 minutes per day, indicating over 30 minutes of use per day among those at the high end of the distribution, alongside a maximum of 172.50 minutes per day.
Additional mean durations of use included 2.23 minutes on weekdays, 2.71 minutes on weekends, 0.71 minutes during school hours, 1.24 minutes after school, and 0.56 minutes at nighttime. Overall, the data indicated notable variability in GenAI use among US youth.
“These findings provide a critical foundation for understanding patterns of GenAI use,” wrote investigators. “Future work must address individual differences, context, and age-related effects of GenAI use.”
References
- Maheux AJ, Akre-Bhide S, Boeldt D, et al. Generative artificial intelligence applications use among US youth. JAMA Netw Open. 2026;9(2):e2556631. doi:10.1001/jamanetworkopen.2025.56631
- Bick A, Blandin A, Deming DJ. The rapid adoption of generative AI. NBER Working Paper. 2025. doi:10.3386/w32966
Newsletter
Access practical, evidence-based guidance to support better care for our youngest patients. Join our email list for the latest clinical updates.






