To address parents’ concerns regarding how AI tools interact with younger users, OpenAI has launched a new safety mode for ChatGPT designed for teens.As the use of AI tools with children and adolescents grows, OpenAI has also added enhanced safety measures and parental controls for teens using ChatGPT. The new ChatGPT for Teens experience will be aimed at teenagers aged 13 to 17, and have more restrictions around conversations with sensitive topics like self-harm, violence, and eating disorders. The shift is part of a larger trend in the tech world of creating an AI system that can react differently to different age groups and requirements.
The new system will be enabled when OpenAI believes that an account is being used by a minor. Teens ages 13 to 17 may also report themselves as being within that age band to avail themselves of the extra safeguards. In instances where OpenAI’s models are not sure of a user’s age, the company says ChatGPT will play it safe, not assuming they are an adult. This will help minimise the risk of younger users accessing interactions which may be inappropriate for them.
One of the biggest differences is the approach that ChatGPT takes to high-risk conversations. However, certain topics involving self-harm, suicide, violence and eating disorders are much more sensitive and call for careful responses as an AI chatbot can have a deeper impact on how a vulnerable user perceives and reacts in a challenging situation. The safeguards for teens aim to put tighter limits on those interactions rather than react in the same way that an adult might.
The shift is crucial, as AI chatbots are no longer relying solely on information-gathering tools. Conversational AI is now being used more frequently by teens for their studies, brainstorming, fun, personal questions and daily conversations. These systems are typically human-like in their natural and conversational style and can make them seem more personable than a standard search engine. This is also helpful if a student requires an explanation, or needs to understand a complicated topic, but can also be risky if a youth starts to use a chatbot to replace trusted human support.
OpenAI’s new protections also feature features that will enable parents to monitor how their teens use ChatGPT. Parents may set up “Quiet Hours” that are designed to restrict access at certain times. They can also handle some account settings and get notifications linked to delicate discussions. Such features enable families to have greater control over how AI is used, while still letting teens use it for everyday tasks, like homework and research or creative projects.
Safety vs privacy issues will probably continue to be a major topic in debates over these controls. Parents might want to be told that their children are safe from exposure to technology that could be harmful, and teenagers might want some privacy when using technology. Creating parental controls that are effective without it being intrusive and making it feel like a parental presence is an intricate endeavor. Features and how they work in real life could be just as significant.
Another aspect of the change is the education part of the teen experience. ChatGPT can be used to navigate students through a problem and to help them understand how to get to the answer, rather than simply giving them answers. This distinction is key in the field of education because it can be easier for the student to finish an assignment if there’s a quick answer rather than learning the concept. A supportive approach can help students to think independently, ask follow-up questions and build problem-solving skills.
But there are concerns with AI education as well, for teachers and parents. Learners can use conversational AI to explain complex topics, practice questions, or delve into topics that go beyond the classroom. Simultaneously, the frivolous use can lead to dependence on produced answers and facilitate academic dishonesty. Effective teaching and parental guidance are essential to solving this problem, but a system that focuses on learning from a teen’s point of view might offer some help.
The announcement follows a broader crackdown on tech firms for the impact their products could have on kids and teens. Newer platforms have become more targets for enhanced safeguards as the government and regulators look into the possible connection between digital platforms and online behaviour with adolescent well-being. Meanwhile, in several countries, concerns have grown about the amount of social media that minors can access, as technology firms are under more pressure than ever to prove that they are making access to social media sites safer for children.
This is a particularly delicate concern for businesses creating AI chatbots, as the systems can have a protracted discussion and communicate in a very personalized manner. A chatbot can seem to listen, recall the direction of a conversation and react with empathy, unlike a traditional website or search engine. If the interaction is with a teenager who is feeling lonely, stressed or in emotional distress, it could be a very personal experience. Supportive technology may be helpful, but there is a distinct difference between providing support and replacing parents, teachers, friends or qualified professionals.
