On August 18, OpenAI officially announced the global rollout of ChatGPT for Teens, a dedicated version designed for users aged 13 to 17, delivering an age-appropriate AI interaction experience. According to OpenAI’s official product announcement, this version retains core AI capabilities while integrating stricter content safety protocols, guided learning modes, and parental control features.
Based on the product details disclosed by OpenAI, ChatGPT for Teens establishes a tiered protection mechanism that is enabled by default. At the content level, the system automatically filters out violent or gory material, sexually explicit content, extreme body-image or beauty standards, fat-shaming, and dangerous challenge behaviors that are unsuitable for minors. Additionally, for high-risk topics such as self-harm, suicidal ideation, and eating disorders, the system proactively triggers intervention measures and pushes safety resources.
On the functionality front, the teen version significantly amplifies its educational focus. OpenAI has partnered with educational organization CodeAI to integrate guided questioning, step-by-step problem-solving, knowledge quizzes, and learning visualization tools into the built-in study mode. When the system detects that a user might be directly seeking homework answers, it gently redirects them toward a comprehension-based learning path, encouraging critical thinking over simply obtaining results. The product also adds more frequent break reminders and a “Quiet Hours” feature, allowing users or parents to set specific time windows when ChatGPT is off-limits, helping to curb overuse.
On the parent side, OpenAI has simultaneously upgraded its family management dashboard, allowing guardians to link their teen’s account and customize settings such as disabling voice conversations, memory features, and image generation modules. Parents can also choose to opt out of having their teen’s conversation data used for model training. All enhanced protection settings are enabled by default; parents may turn them off, but teen users do not have the authority to modify security levels on their own.
Jonas, OpenAI’s head of global policy, told the media in an interview that designing these protective mechanisms requires striking a balance between teen privacy and the risk of harm. As a result, intervention strategies are primarily triggered by clear high-risk signals, avoiding over-intrusion into normal conversation scenarios.
The launch of ChatGPT for Teens marks a “productized consolidation” of OpenAI’s safety capabilities over the past year. In September 2025, OpenAI first introduced parental control features; in November, it released the Teen Safety Blueprint; in December, it added Under-18 Principles to the Model Spec; starting in January 2026, it began rolling out an age prediction system that infers a user’s probable age through behavioral and account-level signals; and on April 23, the GPT-5.5 large model was officially released, with its system card adding a dedicated safety specification chapter on minor protection, clarifying content tiering standards for different age groups. Today’s release of ChatGPT for Teens upgrades this into a full-fledged version for users aged 13 to 17.
Analysts point out that the launch of ChatGPT for Teens is, to a large extent, the culmination of a series of teen safety efforts surrounding chatbot interaction with minors, driven by legal and public pressure. The U.S. Federal Trade Commission (FTC) has previously launched investigations into several AI companies, including OpenAI, focusing on the safety impact of chatbots used as companionship tools for children and teens. The European Union’s AI Act also imposes explicit minor protection requirements on general-purpose AI systems.
A survey released by CodeAI shows that 84% of students are already using AI tools, yet only 16% of high school administrators report that their schools systematically teach AI-related knowledge. Meanwhile, 75% of high school students believe that understanding AI will become increasingly important in the future. This supply-demand gap signals a massive educational market opportunity. At the same time, if teens lack proper guidance when using AI, they face certain potential risks.
Globally, major tech platforms are currently facing unprecedented regulatory and public backlash, with the core controversy revolving around whether platforms profit at the expense of minor users. Coincidentally, today’s launch of ChatGPT for Teens happens to overlap with the start of the Meta social media addiction lawsuit trial.
On August 18, according to CCTV News, the Meta lawsuit jointly filed by the attorneys general of California, Colorado, Kentucky, and New Jersey officially went to trial at the U.S. Federal Court in Oakland, California. The plaintiffs accuse Meta’s social media platforms of addictively designed features that harm minors, seeking damages of up to $1.4 trillion. Meta CEO Mark Zuckerberg and Instagram chief Adam Mosseri are both listed as planned witnesses by the prosecution. That figure is nearly equivalent to Meta’s current total market capitalization, setting a record for the highest damages claim in global consumer protection litigation.
Industry observers widely agree that regardless of the final verdict, this case will become a landmark event in the field of minor protection on digital platforms. The eventual liability determination could very well extend to other digital product sectors such as AI and short-form video, forcing the entire industry to re-examine the ethical boundaries embedded in product design.
