OpenAI has introduced ChatGPT for Teens, a dedicated environment for users aged 13 through 17 that combines educational scaffolding tools with reinforced safety constraints and parental controls.
The rollout automatically routes users into the teen environment if they register as 13 to 17 years old or if OpenAI's automated age-prediction classifier estimates they fall into that demographic. Children under the age of 13 remain prohibited from the platform under OpenAI's standard terms of service.
Learning-First Architecture and Homework Interventions
Rather than functioning as a standard conversational engine, ChatGPT for Teens integrates pedagogical mechanics designed to discourage answer shortcuts. The central feature is Study Mode, developed alongside educational researchers, which uses Socratic scaffolding, metacognitive prompts, and progressive hints rather than returning direct solutions to homework queries.
OpenAI paired Study Mode with several behavioral intervention mechanisms:
- Responsible Homework Reminders: Context-detection routines that flag when a user appears to request immediate assignment answers, prompting the interface to divert the session into step-by-step problem-solving.
- Study Hours Scheduling: Configurable time blocks, managed by teens or linked parent accounts, during which Study Mode is locked on by default.
- Interactive Practice: Built-in quizzes and visualization modules for complex STEM concepts.
Alongside the product changes, OpenAI established a multi-year partnership with education nonprofit CodeAI. The collaboration includes a joint advisory council covering learning science and child development, nationwide classroom curriculum integrations, and the expansion of CodeAI's AI Foundations coursework.
Updated Under-18 Model Spec and Safety Guardrails
Underpinning the release is an updated edition of OpenAI's Under-18 Model Spec. The specification enforces stricter safety thresholds across high-risk domains:
- Content Filtering: Default blocking and real-time intervention for queries involving self-harm, eating disorders, violence, illicit substances, and dangerous activities.
- Relationship Boundaries: Explicit bans on romantic language, simulated emotional dependence, sexualized roleplay, and anthropomorphic affirmations of AI sentience or emotion.
- Parental Supervisions: Linked parent accounts can enforce Quiet Hours, monitor high-level sensitive safety triggers, and control account data retention parameters.
The release consolidates earlier incremental safety pilots, including the age-prediction classifier introduced earlier in 2026 and initial parental control dashboards, into a single unified product surface.
