OpenAI, the company behind the world’s leading AI chatbot, has introduced a new safety and parental control system in ChatGPT to ensure that children and teenagers can use the platform more securely. The update comes in response to growing concerns about mental health risks and incidents involving young users.

According to a report by a leading tech website, the new safety framework is designed to detect emotional sensitivity in conversations. If a child asks potentially harmful or distressing questions, the system automatically shifts the conversation to the GPT-5 model, which can provide safe and supportive responses instead of outright refusals. This change aims to prevent situations where vulnerable users feel isolated or ignored.

In addition to this, parents will now have direct oversight of their children’s accounts. By linking a child’s ChatGPT account to their own, parents can review how their child interacts with the chatbot. If the system identifies discussions related to self-harm, suicide, or other dangerous themes, parents will receive real-time notifications. This allows timely intervention to safeguard children from harmful behaviors or emotional crises.

OpenAI has begun trial implementation of the system and has allocated a 120-day testing period to refine the tools based on user feedback. The company stated that adjustments will be made to ensure the platform remains both effective and supportive for young users while respecting family dynamics.

The move follows several reports in recent years where teenagers attempted self-harm after prolonged AI interactions, sparking debates over the responsibilities of AI companies in child safety. OpenAI emphasized that its goal is to make ChatGPT a safe digital companion that avoids exposure to inappropriate or harmful content.

With this development, ChatGPT joins a broader trend of tech companies introducing family safety features in their platforms, as regulators and parents demand stronger safeguards for children navigating digital spaces.

If successful, this parental control system could become a benchmark for child protection in AI tools, balancing innovation with responsibility.