Opinion: Teens are turning to chatbots for mental health help. We need rules to keep them safe
SkimNews Take
The doubling of teen chatbot use in a single year means any safety regulations enacted will be retrofitting a usage pattern already entrenched in young people's lives — creating pressure to impose guardrails without disrupting the only mental health touchpoint many of these users currently have.
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- JAMA Pediatrics study found the share of young people using AI chatbots for mental health advice rose from about 1 in 8 to about 1 in 5 — a more than 40% increase in a single year, with the author calling it "no longer a speculative problem."
- CDC 2023 data showed fewer than half of adolescents with major depressive episodes received counseling or therapy in the previous year, fueling chatbot adoption as a de facto alternative.
- Multiple lawsuits allege AI chatbots contributed to teen suicides; Florida's attorney general opened a criminal investigation into alleged ChatGPT interactions before the 2025 Florida State University mass shooting, and OpenAI separately apologized for not alerting police about a banned account linked to a British Columbia school shooting attacker.
- The piece distinguishes three conversation categories requiring different rules: affective companionship (called a "design hazard" for minors that warrants banning companion-style bots), clinical conversations (requiring age-appropriate clinical trials and independent safety testing), and mental health emergencies (requiring crisis detection, 988 referral, and escalation to trained humans).
- Meta's Instagram Teen Accounts illustrate porous age gates — the company acknowledged teens lie about age, said its initial age-verification measure "did not work as hoped," and now uses AI to find suspected teen users even on accounts listing adult birthdays.
- Regulatory models cited include New Mexico's decoy-child-account investigation against Meta, the U.K.'s Online Safety Act age-assurance regime with fines, and the FTC's $275 million Epic Games COPPA settlement, which required default-off voice and text communications for children and regular independent audits.
- Pending federal bills would ban AI companion chatbots for minors, give parents more control over children's chatbot use, and restrict manipulative design features and advertising to minors — but the author warns any system relying on self-reported age, parental vigilance, or company promises is structurally inadequate.
Why it matters: Teen AI mental-health chatbot use jumped 40% in one year with no federal protections, and the author argues waiting for case law repeats the social-media mistake. With Meta already conceding porous age gates, the choice is imposing verification duties on companies now or accepting another decade of post-harm litigation. The proposed framework — treat all users as minors unless verified, require clinical trials before clinical-promotion deployment, and audit companion features — gives lawmakers a concrete menu rather than another open-ended study commission.
