Kids & AI · Blog

When something goes wrong online, who does your child tell?

By Ray & Renie Robinson, Aunty Bea  ·  14 October 2026

On 9 October 2026, the Australian Federal Police's Commissioner, Krissy Barrett, put a hard number in front of parents. Of the young people in a national study who said they had been sexually victimised online, 35 per cent either told no one or turned to an AI chatbot for help.

Not a parent. Not a teacher. Not a helpline. A chatbot, or nobody at all.

This isn't a story about one app. It's about where children go when something frightening happens, and how we make sure the answer is still "a person who loves them".

What actually happened

The figures come from AI and non-AI online child sexual victimisation: A national study of Australian Adolescents, by Tim Cubitt and his team at the Australian Cybercrime Observatory at Adelaide University. The research was supported by ICMEC Australia, and the AFP-led Australian Centre to Counter Child Exploitation (ACCCE) and the National Office for Child Safety collaborated on it.

It drew on a nationally representative survey of 1,894 young Australians aged 16 to 18. The study was first released in July. What's new is the AFP putting its weight behind the findings this week, with the Commissioner speaking directly to parents.

Among the seven central findings the AFP listed:

The Commissioner added that more than 15 per cent of young people reported seeking personal or counselling advice from AI at least once a week.

Her message to families was direct: "Parents and caregivers need to know that this is happening." She urged parents to talk with their children about coming forward to a trusted adult, including police, and added: "When someone reaches out for help, particularly about child exploitation or abuse, every minute can matter."

Speaking to the ABC on 8 October, she said: "It starts in the home, it starts with parents knowing what their kids are doing online, creating those safe opportunities for conversations."

Why it matters for Australian families

A chatbot is always awake, never shocked, and never tells Mum. You can see why a frightened teenager might type it out there first. But as the Commissioner put it to the ABC, "AI technology, it's not a human." A chatbot can't make a report. It can't sit with your child. It can't call anyone.

That's the gap. The AFP says it is talking to industry about making chatbots give sound advice, "such as encouraging adolescents to seek help from a trusted adult". That work matters. It doesn't change what happens in your house tonight.

The finding about boys deserves a second read. Boys are more likely to be targeted when AI is involved, and among boys who confided in AI rather than authority figures, two in five told no human at all. If you have a son, don't assume he'll come to you because he hasn't said anything. Silence isn't the same as safe.

Asked whether the under-16 social media ban should extend to chatbots, the Commissioner told the ABC she wouldn't speak to what government should do, and that bans are "not the whole solution". No law sits at the kitchen table. You do. (Which apps the ban currently includes is in our guide to which apps are in Australia's social media ban.)

What you can do tonight

  1. Say the sentence out loud. "If anything ever happens online that scares you or makes you feel ashamed, you can tell me, and you won't be in trouble." Kids don't assume this. They need to hear it, more than once.
  2. Ask about AI without judgement. "Do you ever ask ChatGPT or other AI apps about personal stuff?" If the answer is yes, don't take it away in a panic. Be curious. Then say: "That's fine for some things. For the big things, come to me as well."
  3. Name the other trusted adults. Not every child will tell a parent first. Agree together on two or three people they could tell: an aunt, a coach, a school counsellor.
  4. Know where to report. Online child sexual exploitation can be reported to the ACCCE at accce.gov.au/report. Image-based abuse and serious online harm can be reported to the eSafety Commissioner at esafety.gov.au/report. If a child is in immediate danger, call 000.
  5. Know who your child can call. Kids Helpline is on 1800 55 1800. Write it somewhere they'll see it.

If your child does tell you something, the first words matter most. Stay calm, thank them for telling you, and don't ask them to delete anything or confront anyone. Our guide to sextortion walks through those first steps.

The bigger pattern

This is the second piece of Australian research in as many months pointing the same way. Victimisation clusters in friendship groups, so when one child is affected, their friends may be too. That means the conversation is worth having even if you're confident your own child is fine. They may be the one a friend confides in, and they'll need to know a trusted adult is the next step.

The AFP points families to ThinkUKnow, the ACCCE-led online safety program, as part of the prevention picture.

The most useful thing you can do this week is make sure your child knows you're a safe place to land. Our guide to how to talk to your kid about online safety has the words to start, and there's more at auntybea.app.

Questions parents ask

On 9 October 2026, AFP Commissioner Krissy Barrett said that of the children in an Adelaide University study who reported being sexually victimised online, 35 per cent either told no one or turned to an AI chatbot for help.
Tim Cubitt and the Australian Cybercrime Observatory at Adelaide University, supported by ICMEC Australia, with the AFP-led ACCCE and the National Office for Child Safety collaborating. It surveyed 1,894 young Australians aged 16 to 18.
The research doesn't say that, and neither does the AFP. Its message is about making sure children also come to a trusted adult. A calm conversation about what they use AI for will usually tell you more than a ban.
To the ACCCE at accce.gov.au/report, or to the eSafety Commissioner at esafety.gov.au/report. Call 000 if a child is in immediate danger. Kids Helpline (1800 55 1800) is there for children themselves.

Sources

  1. Australian Federal Police — AFP invests in tech to protect against online harm as teens turn to AI for help — 9 October 2026
  2. ABC News — AFP chief says policing alone cannot fix rise in online child abuse — Olivia Caisley — 8 October 2026
  3. ICMEC Australia — AI- and non-AI online child sexual victimisation, disclosures and help seeking — 6 July 2026
  4. Reporting paths and help lines (VOICE.md §7, the engine's approved list; not re-fetched this run)

Read next: our guide →

Aunty Bea helps Australian families with what comes next, quietly. You'll never scroll your child's messages in Aunty Bea. You get a plain-English summary of what looked risky, and how to talk about it.

Get started free →