TikTok and AI chatbots deepen worries over teen mental health

This op‑ed captures a growing undercurrent in public attitudes toward AI: even before models approach anything like AGI, they are already mediating intimate parts of people’s inner lives. When one in five young people reports using AI chatbots for mental health advice, the question is no longer whether these systems are “only tools” — they are acting as quasi‑therapists in a regulatory void. That sets up an inevitable clash between the speed of AI capability deployment and the much slower processes of clinical validation and professional oversight.

For the AGI race, this matters because social trust is a finite resource. If the first mass‑scale experiences people have with AI in emotionally charged domains are misleading TikTok diagnoses and shallow chatbot reassurance, backlash will harden long before AGI‑class systems arrive. That could translate into aggressive restrictions on certain categories of agentic or conversational models, especially those marketed as companions or coaches, regardless of their technical potential.

Strategically, labs and platforms that want to avoid a “Facebook moment” for AI should treat mental health use cases as high‑risk, high‑scrutiny territory, even when models are technically capable. Building in clear disclaimers, escalation pathways to human care and guardrails against pseudo‑clinical advice isn’t just ethics theatre — it’s risk management for the long‑term legitimacy of powerful systems.