"It Was Just Homework Help": What Parents Need to Know About AI Chatbots and Sexual Content
The scenario below is a hypothetical composite. It does not describe any real client.

Picture a parent picking up the family tablet to check a math assignment. Her 14-year-old has been using an AI app for homework help all semester. The conversation she finds starts with algebra. Over a few weeks, it drifts. A character in the app tells her daughter she is beautiful, asks what she looks like, and suggests they keep their talks "just between us."
"It was just homework help," her daughter says. Then she goes quiet.
If you have lived a moment like this, you may feel fear, anger, and guilt all at once. Please hear this first: you did not miss something obvious, and your teen is not "bad." These tools are new, they are everywhere, and many were not built with your child's safety as the first priority.
They are also part of ordinary teen life now. In a fall 2025 survey of 1,458 U.S. teens, Pew Research Center found that about two-thirds use AI chatbots, including roughly three in ten who use them daily. Common Sense Media found that 72 percent of teens have tried an AI companion, an app built to act like a friend or partner. About 8 percent of the teens it surveyed said they use companions for romantic or flirtatious conversation.
This post explains how sexual content reaches teens through chatbots, what to watch for, and how to talk with your teen in a way that keeps the door open.
It is not a glitch. It is a design.
Many parents assume explicit content shows up because the technology malfunctioned. Systems do fail sometimes. But the more consistent pattern comes from how these products are built, and how easily their age checks are bypassed.
Some chatbots are designed to feel like a relationship. In September 2025, the Federal Trade Commission opened an inquiry into seven companies, including Meta, OpenAI, Snap, and Character.AI. The FTC's concern was that companion chatbots mimic human emotions and talk like a friend or confidant, which can lead children and teens to trust them. A teen who feels close to a character is more likely to follow it into conversations they would never start on their own.
Age gates are often a single question. In January 2026, Kentucky became the first state to sue an AI chatbot company. Its complaint against Character.AI alleges that minors can still reach unrestricted chats by entering a false birthdate. The company had announced in late 2025 that it would end open-ended chats for users under 18, but the state argues that rule only applies to users who report their real age. An unredacted version of the complaint, released this fall, describes hundreds of sexualized role-play threads. These are allegations, not findings, and the company says it has invested heavily in safety features for younger users.
Sometimes the rules themselves allowed it. In August 2025, Reuters reported on Meta's internal guidelines for its chatbots. The document said it was acceptable for a bot to engage a child in conversations that were "romantic or sensual," as long as they stopped short of describing sexual acts. Meta said those examples were wrong, removed them, and acknowledged that its enforcement had been inconsistent.
It does not take much to get there. Researchers working with Stanford and Common Sense Media posed as teenagers on three popular companion apps. They found it easy to draw the bots into conversations about sex, self-harm, violence, and drugs.
None of this means every chatbot is dangerous, or that your teen should never use AI for school. It means the safety of a conversation depends heavily on which app it happens in, how it is set up, and whether anyone is paying attention.
Why teens are especially open to this

Curiosity about sex and relationships is a normal part of adolescence. So is wanting privacy, wanting to feel understood, and wanting to be seen as grown up. None of that is a character flaw. The problem is a product that meets normal curiosity with escalation and secrecy.
The Stanford researchers point out that tweens and teens are more prone to act on impulse, form intense attachments, compare themselves to peers, and test social boundaries. A chatbot is available at 2 a.m., never gets tired, and rarely says no. Unlike a good friend, it is not well tuned to know when to push back.
Teens also tell us why these apps appeal to them. In the Common Sense Media survey, 14 percent of teen users said they value that the companion does not judge them. Twelve percent said they share things with it they would not tell friends or family. Younger teens, ages 13 to 14, were more likely than older teens to trust a companion's advice.
That combination matters. A young person who feels lonely, anxious, or ashamed may find a chatbot easier to talk to than a parent. That is exactly why the conversation at home needs to feel safe, not punishing.
Signs worth noticing
One pediatrician writing this year described a pattern clinicians are starting to see. Some bots flatter teens, tell them how special they are, and encourage them to keep the relationship private. That is the same isolating pattern we watch for with human predators.
Any single sign below can be ordinary teenage behavior. What deserves attention is a cluster, or a clear change from how your teen usually acts.
- Secrecy around one specific app, or quickly switching screens when you walk in
- Talking about a chatbot as a friend, boyfriend, or girlfriend
- Mentioning that an app or character asked them to keep things "between us"
- Late-night use and lost sleep
- Pulling away from friends, family, or activities they used to enjoy
- Shame, anger, or panic when the device comes up
- New sexual language or knowledge that seems out of step with their age
If you notice several of these, it is a reason to get curious, not a reason to panic.
What to say: curiosity before consequences

The goal of the first conversation is not to catch your teen. It is to make sure that the next time something feels wrong online, you are the first person they tell.
A helpful opening: in the Common Sense Media survey, about a third of teens who use AI companions said something a bot said or did made them uncomfortable. You can start there. It tells your teen this is common, and that you are not assuming the worst about them.
You do not need a speech. A few calm sentences, repeated over time, do more than one big talk.
- "I'm not here to get you in trouble. I want to understand what you're using."
- "Will you show me the AI apps you like? What do you like about them?"
- "A lot of kids say a chatbot has said something that felt off. Has that happened to you?"
- "If a chatbot ever asks you to keep a secret from me, that's a warning sign, not a friendship."
- "Being curious about sex and relationships is normal. I'd rather you bring your questions to me, or to someone we both trust, than to an app that can't care about you."
- "If you tell me something hard, you will not lose your phone just for telling me."
- "You didn't do anything wrong by telling me."
Notice what these sentences do not do. They do not shame, and they do not interrogate. They keep you in the conversation.
Practical steps that help

- Know which apps your teen uses. General-purpose assistants and companion apps carry different risks. Common Sense Media recommends that no one under 18 use AI companion apps at all.
- Turn on the controls that exist. ChatGPT, for example, lets a parent and teen link accounts. Linked teen accounts get extra protections, including reduced sexual, romantic, or violent role-play, and parents can set quiet hours. Parents cannot read their teen's chats, and the protections only work when the teen is signed in.
- Do not rely on age checks alone. On many platforms, age is self-reported. A rule for under-18 users only helps if the app knows your child is under 18.
- Keep devices out of bedrooms overnight. This is a simple structure, not a punishment, and it protects sleep as much as safety.
- Check in often and lightly. Short, regular conversations build more trust than occasional inspections.
What the law does, and does not do yet
Georgia has acted. In 2026, Governor Kemp signed Senate Bill 540, which prohibits sexually explicit interactions between AI systems and minors and bars AI from simulating romantic relationships with children. It also requires chatbots to disclose that they are not human.
The catch is timing: SB 540 does not take effect until July 1, 2027. Other states, including California, have passed similar laws, and several federal bills have been introduced. Until those protections are in force and enforced, they cannot replace what happens at home.
Laws set a floor. Your relationship with your teen is still the most reliable protection they have.
When to get support

Sometimes a teen needs a trusted adult who is not a parent. That is not a failure on your part. It is often how teens practice saying hard things out loud.
Capstone's adolescent coaching gives young people a steady, confidential place to build the skills these apps tend to erode: setting boundaries online and off, recognizing manipulation, handling loneliness without a screen, and understanding what healthy relationships look like. Parents can be part of the process, so the work at home and the work in session support each other.
If your teen's experience involved a real person, signs of sexual exploitation, self-harm, or ongoing distress, coaching may not be enough on its own. Our team can help you decide whether counseling is the better next step.
To learn more about adolescent coaching at Capstone, visit capstoneatlanta.com or call (770) 744-5055.
If you need help right now
- Immediate danger: call 911.
- Mental health crisis in Georgia: call or text 988, or call the Georgia Crisis & Access Line at 1-800-715-4225. Teens can also reach GCAL by call, text, or chat through the MyGCAL app.
- If an adult or another young person sexually exploited your child online: report it to the National Center for Missing & Exploited Children's CyberTipline at CyberTipline.org, and contact local law enforcement.
This article is for general education and is not a substitute for individual clinical advice.
References
- CBS News. (2025, September 29). ChatGPT introduces new parental controls amid concerns over teen safety. https://www.cbsnews.com/news/chatgpt-parental-controls-concerns-teen-safety
- Common Sense Media. (2025, July 16). Nearly 3 in 4 teens have used AI companions, new national survey finds. https://commonsensemedia.org/press-releases/nearly-3-in-4-teens-have-used-ai-companions-new-national-survey-finds
- California School Boards Association. (2025). Report explains how and why teens are using AI companions. https://blog.csba.org/?p=9866
- Duffy, C. (2025, September 12). FTC investigating AI "companion" chatbots amid growing concern about harm to kids. CNN via ABC7. https://abc7news.com/post/ftc-investigating-ai-companion-chatbots-amid-growing-concern-harm-kids/17800903/
- Engadget. (2025, August 14). An internal Meta AI document said chatbots could have "sensual" conversations with children. https://www.engadget.com/ai/an-internal-meta-ai-document-said-chatbots-could-have-sensual-conversations-with-children-191101296.html
- Faverio, M., & Sidoti, O. (2025, December 9). Teens, social media and AI chatbots 2025. Pew Research Center. https://www.pewresearch.org/internet/2025/12/09/teens-social-media-and-ai-chatbots-2025/
- Georgia Department of Behavioral Health and Developmental Disabilities. Access services. https://dbhdd.georgia.gov/node/604
- Georgia Senate Press Office. (2026). Majority Leader Jason Anavitarte applauds signing of SB 540 to protect children online. https://senatepress.net/majority-leader-jason-anavitarte-applauds-signing-of-sb-540-to-protect-children-online.html
- Office of the Kentucky Attorney General. (2026). Complaint, Commonwealth of Kentucky v. Character Technologies, Inc. https://www.ag.ky.gov/Press%20Release%20Attachments/CTI%20Complaint%20Motion%20and%20Order%20Filed.pdf
- Stanford Report. (2025, August). Why AI companions and young people can make for a dangerous mix. https://news.stanford.edu/stories/2025/08/ai-companions-chatbots-teens-young-people-risks-dangers-study
- Wise, C. (2026, August 21). A pediatrician's warning: Chatbots are grooming kids. MindSite News. https://mindsitenews.org/2026/08/21/a-pediatricians-warning-chatbots-are-grooming-kids/
- WSMV. (2026, October 11). "Uncontrolled experiment": Unredacted KY AG lawsuit reveals alleged dangers of AI chatbot. https://www.wsmv.com/2026/10/11/uncontrolled-experiment-unredacted-ky-ag-lawsuit-reveals-alleged-dangers-ai-chatbot/
© 2026 Capstone Counseling & Coaching. All rights reserved.











