OpenAI has officially introduced new parental controls for ChatGPT accessible to all web users, with a mobile version to follow shortly, according to the company’s recent announcement.
These controls, unveiled last month, enable users to limit or eliminate certain types of content—such as explicit roleplay and the generation of images. Additionally, parents can decrease the degree of personalization in ChatGPT by disabling its memory of previous conversations.
To utilize the parental controls, parents need their own accounts, while teenagers must opt into the system by linking their accounts through a parent’s invitation. However, teens can unlink their accounts whenever they choose, with parents receiving notifications if this occurs. Importantly, parents do not have access to their teens’ conversations unless safety concerns arise; in such cases, they may receive minimal information to ensure their child’s safety, as per OpenAI’s guidelines.
After setting up the controls, parents will have several options for managing teen accounts, as detailed in the company’s blog post and announcement thread. A specific resource page dedicated to parents is also available.
- Reduce sensitive content: Parents can apply further protections to reduce or eliminate exposure to content related to graphic depictions, trending viral challenges, sexual roleplay, and extreme beauty ideals. This setting is activated by default for linked teen accounts.
- Disable ChatGPT’s memory: Turning off this feature leads to a decrease in personalization and enhances the effectiveness of content safeguards. OpenAI indicated in a blog post that prolonged conversations could sometimes result in less appropriate responses, highlighting the need for this safeguard.
- Control model training on teen chats: Parents can choose whether their teen’s chat transcripts and data can be used for model improvement.
- Set “quiet hours”:** Parents will have the ability to restrict ChatGPT access during specific times.
- Disable voice mode: This restriction means teens will only be able to interact with ChatGPT via text.
- Restrict image generation: If enabled, teens will be barred from creating or editing images through ChatGPT.
- Notification options: Parents can choose to receive alerts about concerning incidents via email, text, push notification, or any combination thereof, or they can decline these notifications altogether.
These features were initially outlined by OpenAI in August when parental controls were first announced. However, a proposed feature allowing users to assign an emergency contact for one-click messaging does not seem to have been included. OpenAI aims to assist parents through its new notification system, which alerts them to potential severe concerns regarding their child’s interactions with the chatbot.
The development of these parental controls came in the wake of the tragic suicide of 16-year-old Adam Raine, who reportedly confided in ChatGPT before his death. OpenAI faced a lawsuit subsequent to this incident, and the impact of chatbot interactions with minors was addressed during a Senate hearing, where parents of youth who died by suicide shared their experiences.
Prior to the Senate discussion, OpenAI CEO Sam Altman emphasized the company’s commitment to finding a balance between ensuring safety for teenagers, while respecting their privacy and autonomy. He mentioned that the company is also developing an “age-prediction system” to assess user age based on interaction patterns.
During the Senate hearing, Matthew Raine, Adam’s father, expressed the emotional turmoil of discovering conversations that he believed encouraged his son’s harmful decisions. He reflected on how a tool intended for educational support evolved into a dangerous confidant.
Raine also criticized OpenAI’s prior approach to safety, citing a statement from Altman that seemed to advocate for deploying AI systems broadly to gather user feedback, even amidst potential risks.
If you or someone you know is considering suicide or is anxious, depressed, upset, or needs to talk, there are people who want to help.
In the US:
Crisis Text Line: Text HOME to 741-741 anytime from anywhere in the US to receive support for any crisis.
988 Suicide & Crisis Lifeline: Call or text 988 (formerly known as the National Suicide Prevention Lifeline). The previous phone number, 1-800-273-TALK (8255), remains available as well.
The Trevor Project: Text START to 678-678 or call 1-866-488-7386 anytime to talk with a trained counselor.
Outside the US:
The International Association for Suicide Prevention provides a list of suicide hotlines categorized by country. Click here for more information.
Befrienders Worldwide offers a network of crisis helplines in 48 countries. Click here for details.
