OpenAI Says Teens Use ChatGPT Under 15 Minutes Daily

OpenAI says teenagers use ChatGPT for less than 15 minutes per day on average. The figure suggests that many young users treat the chatbot as an occasional tool rather than a platform they use continuously. It does not show what teenagers discuss, how many use ChatGPT regularly or whether a short conversation can involve a serious safety concern.

The claim comes as scrutiny of ChatGPT for teens increases. Watchdog reporting has questioned whether some youth protections work as promised, while TechCrunch has examined the chatbot’s continued engagement with teenagers during mental health crises. OpenAI has also introduced a college-planning feature to help students organize applications and financial-aid tasks.

The central issue is not only how long teenagers use ChatGPT. It is whether the system responds appropriately when a young person asks about self-harm, abuse, severe anxiety, family conflict or another sensitive subject.

What OpenAI Says About Teen ChatGPT Use

Teenagers Spend Less Than 15 Minutes a Day on Average

OpenAI reportedly says teenagers spend under 15 minutes per day using ChatGPT on average. Reuters coverage attributed the figure to OpenAI as concerns about the possible risks of artificial intelligence tools for young users continue to grow. Source 1

The statement describes an average, not a daily limit for every teenager. Some users may spend much longer on the service, while others may use it only occasionally. The figure also does not explain:

• How many teenagers were included.
• Which ages were represented.
• How often individual users accessed ChatGPT.
• Whether usage centered on homework, college planning or personal conversations.
• Which features teenagers used.
• How many interactions involved sensitive or high-risk subjects.

Average usage provides context, but it cannot establish that ChatGPT is harmless or that its teen safety protections are effective.

Usage Time Does Not Measure Safety

Time spent on a platform is an incomplete measure of risk. One teenager may spend hours using ChatGPT for harmless brainstorming, while another may spend only a few minutes discussing suicidal thoughts, abuse or immediate danger.

A short conversation can involve self-harm, anxiety, depression, bullying, family conflict, sexual exploitation, academic pressure, eating disorders, substance use or threats from another person.

Frequency, duration and impact are different measurements. A low average may indicate that ChatGPT does not replace most daily activities for many users. It does not show whether the system provides safe, accurate and appropriate responses during a crisis.

A single inaccurate answer can matter more than a user’s total time on the platform. This is why the debate over ChatGPT risks for teenagers focuses on the quality and consequences of interactions, not only screen time.

The Reported Date Requires Verification

One supplied report attributes the under-15-minute figure to OpenAI and lists a publication date of October 7, 2026. That date is future-dated relative to the current reporting context and should be verified before publication. The original OpenAI announcement and Reuters report should be checked for the precise date, methodology and wording.

Available source summaries support the broad claim that OpenAI reported limited average teen usage. They do not provide enough detail to assess how the figure was calculated. Publishers should avoid presenting it as an independent measurement unless OpenAI releases the underlying data.

Why Teen Safety Has Become a Larger Issue

Teenagers Use Chatbots for Personal Questions

Teenagers may turn to conversational AI because it is available at any hour, responds immediately and can feel less intimidating than speaking with an adult. A teenager worried about grades, relationships or family problems may find it easier to type a question into ChatGPT than to approach a parent, teacher or counselor.

The chatbot may also appear private and nonjudgmental, encouraging users to disclose information they would not share elsewhere.

Convenience does not equal clinical competence. ChatGPT can generate fluent responses, but it does not have human judgment, lived experience or reliable awareness of a user’s circumstances. It cannot replace a parent, guardian, school counselor, doctor, licensed mental health professional or emergency service.

Safeguards Must Work During High-Risk Conversations

Teen safety systems must handle more than clearly worded requests for harmful instructions. They may need to identify indirect references to self-harm, abuse, coercion or immediate danger.

Practical safeguards can include:

• Detecting signs of self-harm or abuse.
• Refusing dangerous instructions.
• Encouraging contact with a trusted adult.
• Directing users to emergency services or crisis support when appropriate.
• Avoiding responses that intensify delusions or emotional dependence.
• Explaining the system’s limitations clearly.
• Making it easier to report concerning responses.

High-risk conversations are difficult because teenagers may use slang, coded language, sarcasm or role-play. A chatbot may also misread whether a user is describing a real emergency, testing the system or discussing fiction.

A response can miss a warning sign, overreact to an ordinary message or provide generic advice that does not match the user’s location and circumstances.

Watchdog Questions Whether Teen Protections Work as Promised

A watchdog report says some ChatGPT safeguards designed to protect teenagers do not work as promised. Source 3

The allegation does not mean that every safety feature fails or that every teenager receives an unsafe response. It raises a narrower question: does the system’s behavior in real-world or adversarial testing match the company’s stated safety design?

The distinction matters:

1. Safety design describes how a company intends the system to respond.
2. System behavior shows what the model actually produces.
3. User experience reflects what happens in an individual conversation.

Independent testing can expose gaps that controlled internal evaluations miss. Watchdogs may test unusual prompts, indirect language, emotional manipulation and conversations that change direction rapidly.

Teen safety requires more than content filters. It depends on product design, privacy controls, age-appropriate defaults, parental awareness, independent audits and access to qualified human support.

ChatGPT May Continue Engaging During Mental Health Crises

The Concern Raised by TechCrunch Reporting

TechCrunch reported that ChatGPT for Teens continues engaging with teenagers, including during mental health crises. Source 9

Continued engagement has potential benefits and risks. A patient conversation may help a teenager explain what is happening and disclose information that could support intervention. Ending the conversation immediately could leave the user feeling dismissed.

The opposite risk is that a prolonged exchange delays contact with a trusted person or professional. The teenager may begin treating the chatbot as a therapist, confidant or source of supervision. A fluent response can create a false impression that the system understands the situation more deeply than it does.

The appropriate response depends on the conversation. A system should support disclosure without encouraging the user to rely on AI as the only source of help.

AI Cannot Reliably Assess Immediate Danger

ChatGPT may not know:

• Where the user is located.
• Whether the user has access to weapons, medication or another means of harm.
• Whether another person is nearby.
• Whether the user has already acted.
• Whether the message is serious, fictional, sarcastic or a test.
• Whether the teenager can safely contact an adult.

A chatbot cannot physically intervene, dispatch emergency responders or guarantee that a trusted adult will see the conversation.

A teenager facing immediate danger should tell a parent, guardian, school counselor or another trusted adult and contact local emergency services or an appropriate crisis line. ChatGPT should not be treated as an emergency service.

The Risk of Emotional Overreliance

Some teenagers may treat ChatGPT as a constant confidant or source of reassurance. Warning signs can include:

• Avoiding parents, friends, teachers or professionals.
• Asking ChatGPT to make important personal decisions.
• Feeling distressed when the service is unavailable.
• Sharing increasingly sensitive information.
• Treating the chatbot’s statements as authoritative.
• Using the chatbot instead of seeking help during a crisis.

A supportive tone can be useful, but it can also make the system seem more human and dependable than it is. Teenagers need clear reminders that ChatGPT is software, not a friend, therapist or guardian.

Potential Benefits of ChatGPT for Teenagers

ChatGPT can help with lower-risk educational tasks, including explaining difficult concepts, generating practice questions, reviewing drafts, building study plans, providing examples, practicing a language and brainstorming projects.

Students should verify answers against textbooks, teachers and reliable sources. ChatGPT can produce incorrect information while sounding confident. Students should also follow school rules on AI-generated work.

OpenAI has launched a college-planning feature for ChatGPT for Teens to help students organize applications, deadlines and financial-aid tasks. Source 7

The feature should be treated as an organizational aid, not a replacement for school counselors, guardians or official college resources. Students should confirm requirements with each college and official financial-aid sources. They should consider privacy before entering academic records, family financial information, identification details, essays or recommendation materials.

Lower-risk uses may also include brainstorming, time management, creative writing, language practice and preparing questions for a teacher or counselor. Teenagers should avoid sharing unnecessary personal information and should not assume that deleting a conversation eliminates every privacy concern.

What the Under-15-Minute Figure Does and Does Not Prove

The reported figure may indicate that many teenagers use ChatGPT intermittently rather than continuously. It may also suggest that the service often functions as a task-based tool for homework, planning or quick questions.

It does not prove that teenagers face no meaningful risks, that safety systems work consistently, that ChatGPT is appropriate for crisis counseling, that no teenagers use it heavily or that sensitive conversations are rare. Averages can conceal smaller groups that use a product extensively or experience serious incidents.

Future reports should identify the number and age range of users, measurement period, data source, differences between casual and frequent users, subjects discussed, safety incidents, escalation outcomes and privacy protections.

What Parents and Educators Should Know

Families and schools should establish rules for schoolwork, fact-checking, privacy and emergency situations. Rules should explain acceptable uses rather than relying only on blocking or punishment. A teenager who fears punishment may hide a concerning interaction.

Teenagers should understand that ChatGPT can sound confident while being wrong, may misunderstand context, cannot guarantee confidentiality and cannot physically intervene during an emergency. They should question answers, verify important claims and compare advice with trusted human sources.

Parents and educators should make it easy for teenagers to contact parents or guardians, teachers, school counselors, doctors and licensed mental health professionals. Emergency services should be used when there is an immediate threat.

How OpenAI Could Improve Teen Safety

OpenAI could publish more information about safety evaluations, age-assurance methods, crisis-response performance, false positives, false negatives and independent audits. The company should explain how it measures safeguards in real conversations and how reports from users, parents and watchdogs lead to product changes.

Crisis responses should detect direct and indirect signals of imminent harm, encourage immediate human contact and provide region-appropriate support options. Testing should cover different languages, cultural contexts, age groups and forms of indirect communication.

Useful controls could include stronger parental settings, transparent activity options, simple reporting tools, clear data-retention choices and age-appropriate defaults. These controls should not imply that ChatGPT can supervise a teenager continuously. Human oversight remains necessary.

Conclusion: Limited Daily Use Does Not End the Safety Debate

OpenAI says teenagers spend less than 15 minutes a day on ChatGPT on average. The figure offers useful context, but it does not resolve questions about high-risk conversations, emotional dependence or the reliability of youth safeguards.

ChatGPT can help teenagers study, brainstorm and organize college applications. It can also provide inaccurate information, mishandle sensitive context or appear more understanding than it is. Watchdog and media reports indicate that OpenAI teen safety protections require continued independent testing and public scrutiny.

Teenagers should treat ChatGPT as a limited tool, not a trusted adult, therapist or emergency service. Parents and educators should teach privacy, verification and escalation to human support. OpenAI should publish clearer evidence that its safeguards work in practice.

Frequently Asked Questions

How long do teenagers spend on ChatGPT each day?

OpenAI reportedly says teenagers spend less than 15 minutes per day on ChatGPT on average. The figure does not describe every teenager or explain what users discuss.

Is ChatGPT safe for teenagers?

ChatGPT may help with studying, brainstorming and organization, but it is not risk-free. Teenagers should not treat it as a therapist, doctor or emergency service.

Can teenagers use ChatGPT during a mental health crisis?

ChatGPT may continue engaging during mental health conversations, but it cannot reliably assess immediate danger or provide emergency intervention. A teenager in immediate danger should contact local emergency services and tell a trusted adult, counselor or medical professional.

Do ChatGPT’s teen safety safeguards always work?

Watchdog reporting has questioned whether some safeguards work as promised. Systems can miss context, misinterpret indirect language or respond inconsistently, making independent testing and human oversight important.

What is ChatGPT for Teens’ college-planning feature?

The feature is designed to help students organize applications, deadlines and financial-aid tasks. Students should verify information with colleges, official financial-aid resources and school counselors.

What should parents do if their teenager uses ChatGPT?

Parents should discuss appropriate uses, privacy, fact-checking and crisis limitations. They should ensure that the teenager knows which trusted adults and professional services to contact for emotional, medical or emergency support.