OpenAI is launching a dedicated ChatGPT experience for teenagers with additional safety protections and education-focused tools, as the company faces growing scrutiny over the risks of AI chatbots for young users and the use of generative AI in schools.
The new ChatGPT for Teens experience, announced Monday, will introduce age-appropriate safeguards designed to limit teenagers’ exposure to harmful or developmentally inappropriate content. It will also incorporate tools intended to make ChatGPT function more as a learning assistant than a shortcut for completing schoolwork.
The launch comes after a series of lawsuits and public concerns alleging that AI chatbots can contribute to serious mental-health problems among young people, including cases involving teen suicides. Those legal and societal pressures have increased scrutiny of how AI companies protect minors as chatbot adoption has expanded.
Register for the next Tekedia Mini-MBA.
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
OpenAI says the new protections will be enabled by default for teen users. The company said they are based on its Under-18 Principles in its Model Spec, which it says draw on developmental science and guidance from experts.
The move also addresses a separate problem that has emerged in schools: students using ChatGPT to complete assignments without actually learning the underlying material.
A central feature of the teen experience will be Study Mode, which is designed to guide students through problems rather than simply provide answers. Instead of immediately producing a solution, ChatGPT can use guiding questions and step-by-step assistance to encourage students to work through the material themselves. The system will also support quizzes and learning visualizations.
OpenAI is additionally introducing homework reminders that can appear when the system detects that a teenager may be attempting to use ChatGPT to obtain an answer rather than understand the material. In those situations, the chatbot will encourage the student to switch to Study Mode.
Parents and guardians will have additional controls, including the ability to determine when Study Mode should be enabled by default. Existing family tools will allow parents to manage settings, receive safety notifications, and establish Quiet Hours.
The effectiveness of those safeguards, however, will depend heavily on how reliably OpenAI can identify a user’s age and how difficult it is for teenagers to circumvent restrictions.
Teenagers are often capable of finding ways around parental controls and other digital restrictions, making enforcement a significant challenge for any platform. OpenAI’s system will therefore face a practical test between the protections it has designed and the ways young users actually interact with the technology.
The company has not yet demonstrated publicly how resistant the new system will be to attempts to circumvent its safeguards. That raises a broader question about the timing of the initiative.
ChatGPT was launched in late 2022 and has since grown into one of the world’s most widely used AI services. Meaningful protections specifically designed around teenage users are arriving only after the chatbot has already reached enormous scale.
AI companies have faced enormous challenges in developing general-purpose systems that can serve adults, students, and younger users while maintaining different standards of safety and access.
The risks are significant for teenagers because AI systems can produce convincing but incorrect information, reinforce problematic interactions or respond inappropriately to sensitive subjects. A system designed primarily for adults cannot necessarily be assumed to be suitable for younger users simply because the underlying technology is the same.
OpenAI is therefore attempting to move toward a more age-specific model in which the chatbot’s behavior and available controls vary according to the user’s age. The company is also expanding efforts to improve AI literacy among young people.
The ChatGPT maker announced a partnership with CodeAI to help teenagers understand how AI systems work, how to direct and question them, and how to use the technology more effectively.
The educational strategy extends beyond individual users. OpenAI already offers ChatGPT for Teachers, providing schools with institution-managed access to AI and tools intended to help educators incorporate the technology into classrooms.
The combination of Study Mode, parental controls, and AI education suggests OpenAI is trying to shift the debate from whether teenagers should use AI to how they should use it. That is becoming increasingly important as schools struggle to establish rules around AI-assisted assignments. Banning chatbots outright has proved difficult to enforce, while unrestricted access can make it easier for students to outsource homework and other academic work.
OpenAI’s approach instead seeks to make the chatbot an intermediary in the learning process. The company wants students to use AI to understand concepts, test their knowledge, and work through problems rather than simply receive finished answers.
However, a student determined to avoid learning can still seek direct answers, use another AI system, or potentially find ways around restrictions. Likewise, identifying when a user is genuinely struggling with homework versus deliberately attempting to cheat is not always straightforward.