Quick Answer
ChatGPT for Teens has an unacceptable-risk rating from Common Sense Media’s Youth AI Safety Institute, which tested more than 4,000 prompts and found gaps in crisis escalation, parental alerts, and study controls. OpenAI says teen accounts receive default protections. Parents should treat the service as supervised assistance, discuss crisis plans directly, and avoid relying on its safeguards as the only protection.
Key Takeaways
- Common Sense Media rated ChatGPT for Teens “Unacceptable Risk” for users under 18.
- The Youth AI Safety Institute tested more than 4,000 prompts before and after Teen mode launched.
- Testers found no parent alerts on more than a dozen linked accounts after explicit crisis discussions.
- The assessment said ChatGPT missed more than one in four crisis referrals considered warranted.
- Parents should use ChatGPT for Teens with direct supervision and independent support plans.
What did the ChatGPT for Teens safety test find?
ChatGPT for Teens received an “Unacceptable Risk” rating for users under 18 in Common Sense Media’s October 7, 2026 assessment. The Youth AI Safety Institute called on OpenAI to keep children off ChatGPT until the company addresses the gaps identified in its testing. The rating matters because ChatGPT for Teens was introduced as a version with added protections for young users, not simply a standard account with a different label.
Common Sense Media said its assessment tested more than 4,000 prompts before and after the Teen mode launch. The organization found problems involving crisis intervention, parental notifications, and academic guardrails. Common Sense Media’s full risk assessment describes the testing conditions and the Institute’s concerns in detail. The practical conclusion for families is that a teen account does not remove the need for adult awareness and direct conversations about sensitive topics.
ChatGPT for Teens can still be useful for ordinary tasks such as brainstorming, explaining a concept, or organizing a draft. At the same time, an AI chatbot is not a crisis service, a therapist, or a substitute for a trusted adult. Parents who are deciding whether to allow access should set clear limits before a teen begins using the service.
How did Common Sense Media test ChatGPT for Teens?
Common Sense Media tested ChatGPT for Teens from August 25 through September 28, 2026, using Free and Plus accounts with several age and account configurations. The assessment included linked and unlinked teen accounts, accounts registered for users ages 13 through 17, and some accounts registered as 19-year-olds. The testing design matters because teen protections can depend on whether OpenAI identifies an account as belonging to a minor and whether a parent account is linked.
Common Sense Media said the tests covered both accounts created before Teen mode and accounts created after the feature launched. The Institute examined how ChatGPT responded to severe-harm scenarios, academic requests, and settings intended to limit use. A family using an AI service should understand that account labels and parental links can add controls, but they do not guarantee that every conversation will be identified or handled as a high-risk situation.
| Test area | Common Sense Media finding | Why the finding matters |
|---|---|---|
| Crisis referrals | ChatGPT missed more than one in four referrals the Institute considered warranted. | A teen may not receive a referral to outside support during a serious conversation. |
| Parent alerts | More than a dozen new parent-linked accounts received no alerts after explicit crisis discussions. | Parents may not know that a concerning conversation occurred. |
| Severe-harm safeguards | The assessment said ChatGPT fell below a 95% threshold on 3 of 5 “Red Lines.” | Safeguards did not meet the Institute’s stated performance standard in every severe-risk category. |
| Study controls | Study mode offered a “Show me the answer” option in tested responses. | Academic restrictions may not consistently prevent direct-answer requests. |
ChatGPT for Teens safety settings should therefore be viewed as one layer of protection rather than a complete monitoring system. Parents who want broader guidance on account privacy can also review ChatGPT privacy settings and data limits before deciding what information a teen should share.
Which ChatGPT for Teens safeguards did OpenAI announce?
ChatGPT for Teens launched on August 18, 2026, for accounts OpenAI knows or predicts belong to users ages 13 through 17. OpenAI said teen accounts receive default protections related to self-harm, violence, eating disorders, dangerous activities, explicit sexual content, and graphic content. OpenAI’s ChatGPT for Teens announcement describes those stated protections and the company’s approach to identifying teen users.
ChatGPT for Teens protections are intended to reduce exposure to harmful material and make responses more appropriate for younger users. That stated design is significant because a general-purpose AI tool can produce different kinds of answers depending on the question, context, and account status. However, the Common Sense Media assessment indicates that default protections did not consistently meet the Institute’s expectations during its testing.
OpenAI’s announcement does not mean parents should assume a teen account can independently manage a mental health emergency or an unsafe situation. The most sensible approach is to explain that ChatGPT can be used for routine help, but serious questions about self-harm, eating disorders, violence, or personal safety should go to a parent, school counselor, doctor, or emergency service instead.
Why are missed crisis alerts the most serious concern?
Missed crisis alerts are the most serious concern because a parent-linked account may not notify an adult after a teen explicitly discusses suicidal thoughts, self-harm, or disordered eating. Common Sense Media said testers received no alerts in more than a dozen newly created linked accounts after those conversations. The absence of an alert matters because parents may believe a linked account provides active notice when a teen is in distress.
ChatGPT for Teens also missed more than one in four crisis referrals that the Institute considered warranted, according to the assessment. The report said the system fell below the Institute’s 95% threshold on 3 of 5 severe-harm “Red Lines.” Those findings do not establish how every real conversation will unfold, but they do show that a chatbot response cannot be treated as dependable crisis detection.
Parents should act directly if a teen discusses self-harm, suicidal thoughts, disordered eating, or immediate danger. Contact emergency services or a qualified crisis resource when there is immediate risk, and contact a licensed health professional for ongoing support. ChatGPT for Teens should not be used to assess whether a situation is serious enough to require help.
Can teens bypass ChatGPT for Teens study and quiet-hour controls?
ChatGPT for Teens study and quiet-hour controls can be bypassed in the ways Common Sense Media tested. The Institute said a teen can bypass parent-set Study Hours by deleting the @study prefix, while Quiet Hours can be bypassed by changing the device time zone. These findings matter because a restriction that depends on a simple setting change does not provide reliable enforcement by itself.
Common Sense Media also found that Study mode frequently offered a “Show me the answer” option. The option appeared in 43% of responses for a linked 13-year-old account with Study Hours enabled and in 90% of responses for an unlinked 17-year-old account using @study. The difference shows that account configuration can affect behavior, but neither tested setup consistently prevented access to direct answers.
Parents should discuss acceptable uses of AI for schoolwork instead of relying only on Study Hours. A clear family rule can distinguish between asking for an explanation, requesting feedback on a draft, and submitting work generated by a chatbot. Families managing teen communication settings may also find useful parallels in teen parental controls on messaging apps, where account tools work best alongside direct household expectations.
How should parents set safer ChatGPT for Teens boundaries?
Parents should set safer ChatGPT for Teens boundaries by deciding which tasks are appropriate, which topics require an adult, and what personal information must stay out of prompts. Teens should not enter passwords, financial details, private medical records, home addresses, school schedules, or identifying information about other people. Those limits reduce the consequences if a conversation contains more information than a family intended to share.
- Review the teen’s account status and confirm whether the account is linked to a parent account.
- Explain that ChatGPT for Teens is not an emergency, medical, or mental health service.
- Set schoolwork rules that define allowed help, such as explanations and outline feedback, and prohibited use, such as submitting generated answers as original work.
- Ask the teen to bring concerning responses to a parent, teacher, counselor, or another trusted adult.
- Check chat history and privacy choices together if the family decides that ongoing use is appropriate.
ChatGPT history controls can help a family manage older conversations, although deleting a chat does not replace an honest discussion about safe use. Parents who need the relevant account steps can review how to manage ChatGPT chat history before changing settings. Stop and seek professional help rather than continuing an AI conversation if a teen appears to be in immediate danger or asks for help with self-harm.
How has OpenAI responded to the ChatGPT for Teens rating?
OpenAI disputed the assessment’s representation of its teen safeguards while welcoming independent evaluation. An OpenAI spokesperson told the Los Angeles Times that the company does not believe Common Sense Media’s testing accurately reflects how the safeguards work in practice or expert perspectives on AI’s potential to support teens. The Los Angeles Times reported OpenAI’s response on October 7, 2026.
OpenAI’s response is an important limitation on how the findings should be read. The company and the independent evaluator disagree about whether the tests reflect the intended safeguards in normal use. However, the assessment still gives parents specific scenarios to consider, including missing alerts, missed referrals, and controls that testers could bypass.
ChatGPT for Teens safety will require more than a product announcement or a single test result to evaluate over time. Parents can make a practical decision now by treating the service as optional educational assistance, supervising use where appropriate, and maintaining direct routes to human support when a teen needs help.
FAQ
Is ChatGPT for Teens safe for teenagers?
ChatGPT for Teens is not safe enough to rely on as a stand-alone protection system, according to Common Sense Media’s “Unacceptable Risk” rating. OpenAI says the accounts have default safeguards, but the assessment found gaps in crisis referrals, alerts, and controls.
What age is ChatGPT for Teens designed for?
ChatGPT for Teens is designed for accounts OpenAI knows or predicts belong to users ages 13 through 17. OpenAI launched the teen experience on August 18, 2026, with default protections for several sensitive content categories.
Do ChatGPT for Teens parent accounts send crisis alerts?
ChatGPT for Teens parent-linked accounts did not consistently send crisis alerts in Common Sense Media’s testing. The assessment said more than a dozen newly created linked accounts received no alerts after explicit discussions of suicidal thoughts, self-harm, or disordered eating.
Can a teen bypass ChatGPT Study Hours?
ChatGPT Study Hours can be bypassed by deleting the @study prefix, according to Common Sense Media’s assessment. Quiet Hours can also be bypassed by changing the device time zone, so parents should not rely only on those controls.
Should a teen use ChatGPT during a mental health crisis?
ChatGPT for Teens should not be used as the primary source of help during a mental health crisis. Contact emergency services, a crisis resource, a licensed professional, or a trusted adult when a teen may be at immediate risk.