OpenAI Launches Teen-Friendly ChatGPT with Safety Features and Parental Controls

OpenAI introduces a new version of ChatGPT designed specifically for teenagers, incorporating enhanced safety features and parental controls. The update, launched on Tuesday, aims to provide a safer and more controlled environment for users aged 13-17.

New Safety Measures for Teen Users

The teen-tailored ChatGPT blocks conversations related to suicide, self-harm, and romantic or sexual content. OpenAI has also implemented an age-prediction system to automatically route minors into this mode. Parents can now set quiet hours and receive high-risk safety notifications, ensuring a more secure experience for their children.

Additionally, the platform includes a study mode that encourages educational use by nudging students towards more productive interactions.

Compute Overhead and Monitoring

OpenAI reports that the monitoring overhead for the new teen-friendly version is approximately 20% of the inference compute. This covers all reinforcement learning (RL) training and evaluations involving GPT-5.6 Sol-class and higher models, as well as all inference on the Astra model.

Industry Context and Implications

The launch of the teen-friendly ChatGPT reflects a growing trend in the tech industry to create safer and more controlled environments for younger users. As AI technology becomes increasingly integrated into daily life, companies are taking proactive steps to address concerns about the impact of these tools on young people.

Experts believe that these measures will help mitigate potential risks and foster a more positive and educational experience for teens using AI chatbots.

References

• AI News Today, August 21 — Top AI Stories & Live Updates | AI Weekly