The most important aspect of the new ChatGPT for Teens feature will probably also annoy teenagers the most. If they ask it to do their homework for them, the chatbot is supposed to recognize it, and will instead steer the student toward figuring it out themselves rather than produce a ready-to-paste essay.
That sounds like a small adjustment to the usual ChatGPT formula, but it gets at one of the biggest problems with allowing students easy access to AI chatbots. While ChatGPT’s abilities as a tutor, research assistant, and endlessly patient explainer are a potential boon to learning, letting it finish a student’s homework while they scroll TikTok is something everyone should want to prevent.
ChatGPT for Teens is one way OpenAI is trying to steer away from that depressing scenario. The experience is intended for users aged 13 to 17 and is automatically applied when someone gives an age in that range or OpenAI’s systems estimate that the user is under 18.
The Responsible Homework Reminder might actually bring some positives to the idea of teens using ChatGPT, not just ways of limiting harm. The feature supposedly recognizes when a teenager wants the AI to do their homework and redirects them toward Study Mode.
Questions instead of answers
There has always been a slightly absurd tension around AI and education. We want students to have access to technology that can explain almost anything on demand, but preferably without allowing that technology to do all the learning for them. Giving a teenager access to ChatGPT and assuming they’ll always use it ethically and to help them learn is absurd. Even with the best intentions, the AI is usually slanted toward being overly helpful and undercutting the kind of education real research and effort provide.
OpenAI takes an engineering approach to a solution. ChatGPT for Teens combines Study Mode and its learning-centered conversations, quizzes, and now homework reminders. Teens and parents can also set Study Hours, making Study Mode the default during those times.
As Study Mode is built around guiding questions and mimicking educators, it should at least reduce the incidence of kids relying on ChatGPT to do the work for them. OpenAI says the system was developed with teachers and scientists to encourage active participation.
And while simply banning ChatGPT as much as possible sounds simpler, there’s a reason some educators prefer solutions like ChatGPT for Teens. A teenager stuck on an algebra problem at 10:30 at night or who doesn’t have the resources to hire a tutor can at least ask ChatGPT to help them work through the problem, and with Study Mode, that means more than just getting the answer.
“Not every student has someone at home who can help when they get stuck or have questions with their homework. With the right guardrails, AI can help fill that gap,” Racquel Gibson, a high school math teacher in Miami, said in a statement. “Tools like Responsible Homework Reminder can guide students step by step, helping them understand their mistakes, build confidence, and keep learning — not simply hand them an answer. That kind of individualized support can make a meaningful difference for students who want to succeed but don’t always know how to move forward.”
There is an important caveat. Guardrails only matter if they actually hold up when a determined teenager starts rattling them. The age prediction operating behind the scenes should supposedly work out when a teen is using the AI. OpenAI says it can use signals including account age, usage patterns, topics discussed, and the times of day ChatGPT is used to estimate whether someone is under 18. OpenAI has admitted it’s not a perfect system, but that it’s still working to improve safeguards.
Teen safety
ChatGPT for Teens also includes reinforced guardrails to identify and prevent conversations around topics like self-harm and other violence, eating disorders, and sexual content. Those guardrails are arguably even more important than the ones stopping kids from using the AI to do their homework.
Even so, concerns about misinformation, chatbot addiction, and an overall degradation of thinking caused by AI chatbots are justifiably on the minds of many parents.
But at least the Responsible Homework Reminders are a gesture in the right direction, recognizing that the AI does not have to maximize how much work it performs on your behalf to be useful. For a teenager trying to learn, doing slightly less can occasionally be the more sophisticated feature.
If OpenAI can make that behavior reliable, teenagers may complain that ChatGPT has become less cooperative just when an assignment is due. But being less helpful might actually make ChatGPT much more useful in the long run.
