A child’s digital life is not one behavior. Chatting with a friend, completing homework, watching a video, and asking an AI for help can have different purposes and risks.
Protect without flattening the picture
The U.S. Surgeon General’s advisory describes meaningful risks and possible benefits of social media for young people. The National Academies’ summary says the evidence did not establish population-level causation and calls for both protections and better research. These conclusions can coexist: uncertainty about average effects does not erase a particular child’s difficulty.
Make learning visible
When a school uses AI, the valuable output is not a polished paragraph on a screen. It is a student who can explain, question, revise, and apply an idea. UNESCO’s guidance and competency frameworks emphasize human-centered use, critical thinking, and teacher judgment.
A classroom exercise
Ask students to propose an answer before consulting AI. Let the system produce a different explanation. Then have students compare both with a source, identify what changed their minds, and describe what remains unknown. The AI becomes a conversation partner whose work can be inspected, not an authority that replaces learning.
Ask what a child is doing online, with whom, and how it feels afterward. Protect sleep, privacy, and time together. Avoid treating every screen use as identical or asking a child to solve a platform-design problem alone.
