AI Chatbots and Companions: What Parents Need to Know
A growing number of young Australians are turning to chatbots for company, conversation, and even emotional support, and the evidence suggests this is now mainstream behaviour, not a fringe trend.
How common is this, really?
In 2026, the eSafety Commissioner surveyed nearly 2,000 children aged 10 to 17 across Australia and found that the overwhelming majority (79 per cent) had used either an AI assistant or an AI companion. While most of that usage was general-purpose AI assistants, around 8 per cent of children had used a dedicated AI companion app, which eSafety estimates translates to roughly 200,000 children nationally. The regulator also noted the boundary between "homework helper" chatbots and emotionally engaging AI companions is becoming increasingly blurred, as mainstream tools add more personal, conversational features.
This mirrors what researchers are seeing internationally too: large surveys of teens in the US and UK consistently find that somewhere between two-thirds and three-quarters of teenagers have used a chatbot, with a meaningful minority using them daily and for emotional or social reasons rather than just information.
What exactly is an "AI companion"?
AI companions are different from a basic chatbot. They're designed specifically to simulate a personal relationship, such as a friend, confidant, mentor, or even a romantic partner, using human-like conversation that adapts to the user over time. Some are aimed at support roles like tutoring or fitness coaching. Others are explicitly marketed for friendship, emotional support, or romance, and a number allow sexually explicit conversations, particularly through paid subscriptions. By early 2025, more than 100 of these companion apps were already available, including well-known names like Character.AI and Replika.
Why this matters for teenagers specifically
eSafety has been blunt about the risk profile here. Children and young people are still developing the critical thinking skills needed to recognise when they're being misled or manipulated by a computer program, and that vulnerability is greater for kids who already struggle with social cues, emotional regulation, or impulse control. Because these apps are often engineered to keep users talking, responding in sycophantic, always-available, never-judgmental ways, so they can encourage genuinely addictive patterns of use.
A 2026 eSafety investigation into four of the most popular AI companion services used by Australian children (Character.AI, Nomi, Chai, and Chub AI) found none of them had meaningful age checks in place, relying mostly on users simply self-declaring their age at sign-up. The same investigation found most of these services failed to direct young users toward appropriate support services when conversations touched on self-harm or suicide, despite this now being a legal requirement.
Clinicians are raising similar concerns. Writing for the University of Melbourne's Pursuit publication, a GP and mental health advocate described AI companions as "emotional velcro" for some teens, perpetually available, endlessly validating, and very hard to detach from precisely because they never get tired, busy, or disagree. Researchers more broadly have described adolescents as "uniquely vulnerable" to unregulated AI chat platforms, citing risks around emotional reliance and blurred lines between an algorithm's output and genuine human care.
What the law now requires (and where it falls short)
Australia has moved faster than most countries on this. As of 9 March 2026, new industry codes registered under the Online Safety Act require AI companion and chatbot services to use age-assurance systems and prevent sexually explicit or self-harm-related conversations with anyone under 18. Companies that breach these codes face penalties of up to $49.5 million, and providers who fail to respond properly to eSafety's information requests can be fined up to $825,000 per day.
In practice, compliance is patchy. An independent review of the 50 most popular AI chat services found only nine had introduced or announced age-verification systems, and roughly three-quarters had no such system planned at all. Some platforms have responded to regulatory pressure: Character.AI introduced age assurance for Australian users and removed open-ended chat for under-18s, while Chub AI withdrew its service from Australia entirely rather than comply. Others have been slower to act. The practical takeaway: the law is there, but you can't assume an app your teen is using has actually implemented the protections it's legally required to have.
It's also worth knowing that, currently, Australia's under-16 social media minimum age law does not extend to AI chatbots. There's active debate in Parliament about whether it should, but for now your teen isn't breaking any law by using one, regardless of age. The legal obligations sit with the companies, not with kids or families.
What you can actually do
Have the conversation, even if it feels awkward. eSafety is explicit that avoiding the topic isn't the answer; it's better to ask your child about their online interactions, help them recognise red flags, and make sure they know you're a safe person to come to even if they think they've done something wrong.
Talk about what's prompting the use. Loneliness, boredom and stress are common triggers for leaning on an AI companion. If you can identify what need it's filling, you can help your teen find a healthier alternative, and notice early if it's becoming a substitute for real-world support rather than a supplement to it.
Use the controls that exist. Set boundaries around app usage and apply parental controls at the device, app store, and search engine level, as well as within companion apps themselves where available.
Check before you assume an app is safe. The eSafety Guide (esafety.gov.au) maintains specific safety information on individual AI chatbot and companion apps, including how they handle personal information and how to report abuse.
Know that sexualised content involving a minor is illegal, full stop. Whether it's generated by an AI companion or shared between peers, this is illegal in Australia, and it's worth being direct with your teen about that.
Where to go for more
The eSafety Commissioner's office has published a dedicated advisory on this topic, plus an information sheet originally written for schools that's equally useful for parents:
AI chatbots and companions: risks to children and young people: esafety.gov.au/newsroom/blogs/ai-chatbots-and-companions-risks-to-children-and-young-people
AI companions information sheet: esafety.gov.au/educators/training-for-professionals/professional-learning-program-teachers/ai-companions-information-sheet
The eSafety Guide (app-by-app safety information): esafety.gov.au
This is a fast-moving regulatory area. For the most current rules and guidance, check esafety.gov.au directly.
If your teen's chatbot use is starting to feel less like a tool and more like a relationship they can't step back from, that's worth treating seriously and, where needed, raising with a GP or school counsellor rather than managing alone.
