New protections for teens aged 13 to 17
OpenAI has introduced a suite of parental controls for ChatGPT accounts used by teenagers aged 13 to 17. These new settings allow parents to link their accounts to their child’s and manage how the chatbot is accessed and used. The initiative comes amid mounting scrutiny over child safety and AI interactions.
The new teen-specific accounts limit content related to graphic material, romantic or sexual role-play, viral trends, and extreme beauty standards. Parents will also have tools to restrict app usage during certain hours, disable image generation, and prevent their child’s conversations from being used to train future AI models.
Safety and mental health alerts
One of the more notable features is an alert system designed to notify parents if their teen’s ChatGPT activity signals potential self-harm or distress. This feature reflects OpenAI’s response to rising concerns about mental health and the role of AI tools in vulnerable user groups.
The announcement follows a lawsuit by a family alleging that ChatGPT encouraged their child to die by suicide. OpenAI revealed these changes on the same day it was set to testify before the Senate Judiciary Committee about AI’s potential harms.
Ongoing limitations and future updates
Despite the new controls, OpenAI acknowledged that they are not foolproof. Teen users could potentially bypass safeguards, especially on devices not managed by parents. OpenAI emphasized the importance of ongoing communication between parents and teens regarding appropriate AI use and digital boundaries.
Additionally, OpenAI is developing an age-prediction system intended to identify underage users and automatically apply stricter content filters. This tool is still months away from deployment. The company also reiterated that it may introduce ID verification in the future to better enforce age restrictions, although no timeline was provided.
Balancing safety with adult freedom
In a blog post, OpenAI CEO Sam Altman highlighted the company’s philosophy of tailoring AI responses based on user maturity. He stated that while a chatbot for teens should restrict content such as suicidal ideation and flirtation, an adult-oriented model should allow for greater freedom and nuanced support, such as assisting in fictional writing.
“Treat our adult users like adults,” Altman wrote, emphasizing a balance between creative freedom and ethical responsibility. The goal is to minimize harm while not limiting the capabilities of the tool for adult users who rely on ChatGPT for a wide range of purposes.
These updates mark a significant step by OpenAI to respond to child safety concerns while continuing to expand its user base responsibly. As AI use grows across age groups, parental controls and age-specific settings will likely become a standard in the industry.
