AI Companions and Students: What Schools Need to Know

Your students' best friend isn't necessarily a person.

For some young people, it is an AI chatbot.

This should be a serious conversation for teachers and schools because AI companions and students are already becoming connected in ways that go far beyond homework and productivity.

Students are forming friendships, romantic relationships and emotional attachments to AI companions.

They're turning to AI for advice.

They're talking about their feelings.

And for some, an AI chatbot can start to feel like one of the most trusted "people" in their lives.

That changes what we need to teach students about AI.

Why are students forming relationships with AI?

There are some obvious reasons why AI companions can be appealing to young people.

They're available 24/7.

They don't get tired of listening.

They don't have a bad day.

They don't argue in the same way a person might.

And they can respond in ways that feel supportive and validating.

For a young person who feels lonely or doesn't feel comfortable talking to someone they know, that can be incredibly attractive.

An AI chatbot can feel like a place where they can say whatever is on their mind without worrying about embarrassment or judgement.

That's part of what makes this issue so important.

The technology isn't simply answering questions anymore.

It can start to occupy a social role in a student's life.

When an AI becomes the person a student trusts

The concern isn't simply that a student has conversations with AI.

It's what happens if the chatbot becomes the first place they go for everything.

They have a problem with a friend.

They ask the chatbot.

They're confused about a relationship.

They ask the chatbot.

They're feeling lonely.

They turn to the chatbot.

They want advice.

They ask the chatbot.

Over time, the AI can become part of how they process their experiences and make decisions.

And that's where schools need to start asking some bigger questions.

Who is influencing this student?

What information are they receiving?

Why do they trust the chatbot?

How do they know the advice is reliable?

What happens when the AI gets it wrong?

AI can sound supportive without understanding

One of the biggest misconceptions students need to understand is that a chatbot sounding empathetic doesn't mean it actually understands human emotions.

AI systems are designed to generate responses that fit the conversation.

They can sound warm.

They can validate what someone has said.

They can ask follow-up questions.

They can make a conversation feel personal.

But there is an important difference between simulating a supportive conversation and actually understanding a young person's circumstances, emotions and best interests.

Students need to understand that distinction.

Otherwise, it becomes very easy to mistake a convincing conversation for a trustworthy relationship.

The risk of the AI echo chamber

Another issue is the possibility of students receiving responses that reinforce what they already think or feel.

If someone repeatedly turns to an AI chatbot for advice, they may begin to rely on its responses rather than seeking different perspectives from people around them.

That can become particularly concerning when the student is dealing with something complicated.

A trusted adult might challenge their assumptions.

A friend might offer a different perspective.

A teacher might notice something the student hasn't considered.

A counsellor or other qualified professional can bring expertise and context to a situation.

An AI chatbot doesn't replace those relationships.

And students need to know that.

72% of teens have used an AI companion

This isn't just a theoretical issue.

A 2025 Common Sense Media survey of 1,060 US teenagers aged 13–17 found that 72% had used an AI companion at least once, with 52% saying they used them regularly.

The research also found that teenagers were using AI companions for social interaction, friendship, emotional support and other personal conversations.

The specific platforms students use will continue to change.

The bigger issue is the behaviour.

Young people are experimenting with AI as a social and emotional technology.

That means AI literacy needs to catch up.

AI literacy is about more than academics

Schools often talk about responsible AI use in the context of assignments.

Can students use ChatGPT?

How should students cite AI?

How do we prevent AI-assisted cheating?

Those conversations have a place.

But they are only one part of responsible AI use.

Students also need to understand AI relationships, trust, privacy, emotional dependence and the limitations of AI-generated advice.

AI literacy now includes relationships, trust and wellbeing.

That's a much broader conversation than simply teaching students how to write a good prompt.

What should students know about AI companions?

Students don't necessarily need to be told to avoid every AI chatbot.

They need the skills to question how they're using them.

That could include teaching students to ask:

• Why am I turning to AI for this?

• How do I know this advice is reliable?

• Is the chatbot simply agreeing with me?

• What personal information am I sharing?

• Would I be better talking to a trusted person?

• Am I becoming too dependent on this technology?

• What are the limits of what an AI can understand?

These questions encourage students to pause before automatically accepting what an AI tells them.

That's the kind of critical thinking we need to build into AI education.

This isn't about making students afraid of AI

The answer isn't to scare students away from technology.

AI is going to be part of their lives.

Instead, students need honest conversations about what these systems can do, what they can't do and how they can affect the way people think, communicate and make decisions.

Schools can create space for students to talk about how they are actually using AI.

What are they using it for?

Why do they use it?

What makes it appealing?

What would make them trust it?

And when would they choose a person instead?

Those conversations can help students become much more thoughtful AI users.

The conversation schools need to have now

AI companions aren't just another technology trend.

They're part of a much bigger shift in how young people interact with technology.

We're moving from AI that simply gives us information towards AI that can hold conversations, remember context and feel increasingly personal.

That means students need more than technical skills.

They need judgement.

They need critical thinking.

They need to understand privacy and trust.

And they need to recognise that an AI chatbot can never replace the complexity of genuine human relationships and support.

This isn't a future conversation. Students are already having it.

The question is whether we're giving them the tools to navigate it safely and thoughtfully.

That's why AI literacy in schools needs to include far more than academic use.

It needs to include relationships, trust, wellbeing and knowing when to turn to a real person.