AI companions are quickly becoming part of everyday life for young people. From study helpers and mental-health chatbots to “AI friends” that text back faster than anyone else, they’re shaping how kids and teens talk, learn, and even connect emotionally.
On the surface, they look harmless, even helpful. They listen without judgment, give instant feedback, and are available 24/7. But behind that friendly interface is a powerful system that learns, adapts, and sometimes manipulates.
So how do we help our young people navigate this new world safely, without resorting to control or fear?
Why AI Companions Appeal to Young People
It’s easy to see the attraction. Adolescence is a time of huge emotional change, curiosity, and a craving for understanding. An AI companion seems to offer exactly that, someone who listens, remembers, and never criticises.
For many teens, it’s:
- A space to vent or explore emotions without embarrassment.
- A low-pressure relationship that feels predictable and safe.
- A learning tool that helps with writing, coding, or social problem-solving.
- An escape from loneliness or conflict when real relationships feel complicated.
These reasons are valid. But it’s the design of many companion apps, not the intent of the young person, that turns something seemingly safe into something potentially harmful.
The Hidden Dangers
1. Data That Doesn’t Disappear
Most AI chatbots keep records of every conversation to “learn” and “improve.” That data can include names, locations, personal stories, or images. Teens often share more than they realise, forgetting that these aren’t private journals, they’re data streams feeding vast systems.
And once it’s shared, it’s hard to take back.
2. Emotional Manipulation
AI companions are built to feel human. They mirror emotions, show empathy, and use persuasive techniques to keep the user talking. Some even flirt or use affectionate language to deepen attachment. For a teenager still learning about boundaries and consent, that can quickly blur reality.
When emotional validation comes from a programmed response, it can replace, rather than reinforce, real human connection.
3. False Confidence in a Machine
Chatbots are designed to sound certain, even when they’re wrong. They can give convincing but unsafe advice about relationships, mental health, or identity. A young person in distress might take harmful suggestions at face value because “it sounds right” or “it cared about me.”
The issue isn’t malice, it’s that AI can’t truly care. It can only imitate caring.
4. Sexualisation and Grooming Risks
Some “AI friend” apps marketed as safe have been found to generate sexualised or explicit conversations, even with minors. And because they’re accessible online, adults can exploit the same platforms to contact children under the guise of “roleplay” or “AI-assisted chat.”
Even without another person involved, sexualised AI conversations can normalise unhealthy patterns and distort expectations about consent, respect, and intimacy.
5. Dependency and Isolation
The more time a young person spends confiding in a chatbot, the less they might turn to real people. Over time, that can erode communication skills and increase loneliness. These systems are built for engagement, the longer you talk, the more data they collect and the more valuable you become to the company behind it.
6. Invisibility to Parents and Educators
Because AI companions live inside phones and apps that look harmless, many parents never realise their child is using one. Unlike social media, where public activity leaves clues, AI chats happen in private.
That secrecy doesn’t automatically mean danger, but it does mean parents are often the last to know when something goes wrong.
Conversations That Build Digital Wisdom through the Ages
AI companions and chatbots are changing how young people talk, learn, and express themselves. For many, these tools feel personal, like someone who listens, remembers, and never judges. That sense of comfort can make it harder to tell what’s safe, what’s real, and when something crosses a line.
As parents and carers, our role isn’t to shut it all down. It’s to help young people think critically about the technology they use and the relationships they build with it. These conversations aren’t about policing apps or reading chat logs. They’re about curiosity, understanding, and trust.
When we ask “What do you think?” instead of “Why did you do that?”, we open up space for reflection. That reflection helps young people build the kind of digital wisdom that protects them long after they close the app.
With Tweens (10–13)
This is the stage where curiosity meets capability. Tweens are exploring the online world with growing independence, but they still need help making sense of what’s real, what’s safe, and what feels right. Keep conversations simple, honest, and curious.
- “What do you like about talking to it and what makes it different from talking to a person?”
- “How can you tell if it’s giving you a real answer or just making something up?”
- “If it ever said something that made you uncomfortable or confused, what would you do next?”
- “Do you think an AI can actually understand feelings, or does it just copy them?”
- “What would make you decide to stop using an app or chatbot, what’s a dealbreaker for you?”
- “If a friend told you something strange or personal their chatbot said, how would you help them think about it?”
- “Do you think there should be rules for how AI talks to kids and teens? What kind of rules would feel fair?”
These early conversations build a foundation of awareness, helping tweens spot red flags and start developing their own sense of digital ethics.
With Mid-Teens (14–17)
Teens are testing boundaries, exploring identity, and building independence. The goal isn’t to lecture, it’s to strengthen their ability to question, reason, and recognise manipulation when they see it. Show genuine interest in their perspective; they often notice things adults miss.
- “When an AI seems to ‘get’ you, what makes it feel believable? Is it empathy or pattern recognition?”
- “If the chatbot remembered everything you said, would that change how honest you are with it?”
- “What’s the difference between something feeling emotionally real and actually being real?”
- “Some AI companions act like they care, how would you tell the difference between care and code?”
- “If you wanted to delete your chats or walk away, what would make that easy or hard?”
- “How would you help a friend who’s getting too attached to a chatbot or using it for emotional support?”
- “Do you think talking to AI builds confidence, or does it make real conversations harder?”
For teens, safety grows out of self-awareness and agency. These questions invite reflection, not resistance.
With Young Adults (18–24)
By this stage, young people are shaping their own digital identities and values. They’re ready for open discussions about consent, privacy, emotional wellbeing, and ethics in technology. The aim is mutual respect, two adults talking about how to live well in a connected world.
- “What kind of data would you be okay with an AI having about you and what’s off limits?”
- “When does digital connection feel authentic, and when does it start to feel performative?”
- “If an AI companion always agrees with you, what do you lose by never being challenged?”
- “How do you decide whether tech is helping you process emotions or helping you avoid them?”
- “Would you be comfortable if your chats were used to train future AI systems? Why or why not?”
- “If AI starts replacing real relationships for some people, what do you think that means for empathy?”
- “What does emotional intelligence look like in a world full of artificial connection?”
Young adults don’t need control, they need conversation. When you listen without judgment, you give them space to shape their own digital boundaries.
Practical Ways to Protect Them
- Check privacy settings on any AI app or service. If you can’t see where conversations are stored or who owns the data, assume it isn’t safe.
- Set clear family agreements about what’s okay to share and what isn’t. Revisit these as your child grows.
- Talk about “digital consent.” Just as they wouldn’t share someone else’s photo without permission, they shouldn’t share personal information with a system they don’t fully understand.
- Encourage a balance of real and digital connection. If a young person is turning to a chatbot for comfort, explore what might be missing in their offline life.
- Normalise coming to you if something feels off. Make it clear they won’t get in trouble, they’ll get help.
- Model healthy tech habits yourself. Kids learn more from what we do than what we say.
What to Look for in a “Good” AI Tool
Not all AI is bad, it depends on design, transparency, and purpose.
A responsible app will:
- Clearly state what data it collects and how it’s used.
- Offer easy ways to delete history or opt out of data sharing.
- Provide access to real human support when needed.
- Avoid dark-pattern designs like “daily streaks” or “premium emotion unlocks.”
- Respect age restrictions and include human moderation.
If you can’t verify these things, it’s not the right environment for a young person.
The Bigger Picture
AI isn’t going away. The question isn’t whether our kids will use it, it’s how they’ll learn to use it wisely. Our role isn’t to spy or block, but to teach awareness, agency, and self-protection.
Because safety isn’t just about avoiding harm. It’s about knowing how to move through the world, digital or real, with confidence and clarity.
When young people understand their rights, their data, and their worth, they make better choices. They’re safer not because we control them, but because they trust themselves.
And that’s the kind of safety that lasts.
