The Bots & The Bees: Stranger danger in the AI era

Most parents have some version of the stranger danger talk memorised: don’t share your name, don’t share your location, don’t talk to people you don’t know. It’s a good foundation. It’s just not built to suit the places children are encountering strangers today.

According to the new Norton Insights Report: Connected Kids, gaming is where exposure is concentrated, not social media. Among the 46% of Australian parents with a child who plays chat-enabled online games:

• 49% say their child has chatted with a stranger via voice or text chat.
• 24% say their child has received inappropriate or offensive messages.
• 24% say their child has been encouraged to leave the game and continue the conversation on a different platform.

Compare that to how parents rank their worries: TikTok tops the list of platform concerns at 21%, with online gaming down at just 9%, behind Facebook (10%) and YouTube (6%). That ranking highlights that social media bans may miss the mark.

Why gaming chat is the blind spot for parents

The internet parents grew up with has changed. In gaming chat, there may be no profile to check and no obvious red flags: just a voice in a headset or a message in a lobby. And the risk isn’t only who’s in the chat. It’s what happens when a conversation moves somewhere else.

Norton’s Global Head of Scam Research, Leyla Bilge, puts it this way: “The chat layer on top of gaming is often where the real exposure happens, and it tends to be less visible to parents than social media use. When a child moves from a game platform to a private messaging app because someone asked them to, that escalation can be hard to detect, and kids may not recognize it as something worth mentioning.”

A move from an in-game chat to Discord or Snapchat can look, to a child, like making a new friend faster. But this is also a documented grooming pattern. Children may not recognise the request as a warning sign, so parents should explain directly that being asked to move a conversation off-platform is always worth mentioning.

Online strangers aren't always people

The new Norton Insights Report also surfaces a second shift: some of the accounts kids are talking to aren’t run by people at all. Among parents whose child has experienced cyberbullying, 30% say the perpetrator was an AI-generated account or bot, slightly higher than the 27% who say it was someone using a fake profile.

That changes the shape of the risk. A conflict with a classmate has context a child can reason through. A harassing message from an account that may not be human doesn’t. There’s no motive to work out and no relationship to repair, just a pattern that keeps escalating.

How to update the "stranger danger" talk

Updating the talk doesn’t mean a bigger, scarier lecture. It means being specific about the situations that children are in. Here are some tips for talking to your child about strangers online:

• Name the platform-hop as a rule, not a judgment call: Tell your kid directly: if someone asks to move a conversation from a game to another app, that’s something to mention to you, no matter how casual it seemed. A clear rule removes guesswork.
• Ask about chat, not just apps: “Anything weird happen in chat this week?” surfaces more than “how was Roblox?” because it’s specific to where the risk lives.
• Talk about bots as a category, not a scare tactic: Children don’t need to be afraid of every message. They need to understand that some accounts aren’t people, and that persistent or manipulative messages are worth flagging no matter what’s behind them.
• Make it ongoing: Just under half of parents surveyed who found out their child had been cyberbullied learned about it because their child told them directly. A single talk won’t close that gap. Regular, low-key check-ins will.

The Digital Wellness Lab, a research center at Boston Children’s Hospital, uses the Five M’s of Digital Wellness framework: Model, Mentor, Monitor, Mastery, and Meaning. For stranger contact specifically, the “meaning” conversation may matter most.

Parents can ask their children what they think someone online actually wants from them, not just whether the person seems nice. Kids are good at reading friendliness, but they’re less practiced at reading intent, especially when friendliness is part of the strategy.

Get more visibility into your kids' online activity

Regular conversations are important, but it helps to know where your child is spending time and who (or what) they may encounter there. Norton Family gives parents visibility into where kids are spending time online, including activity across devices and apps, so these conversations can be informed by more than a guess.

Methodology note

*The 2026 Connected Kids study was commissioned by Gen (the company behind Norton) and conducted online by Dynata across 12 markets from 22 May to 19 June 2026. It included 1,000 Australian adults. Data is weighted by age, gender, and region to be nationally representative. Mentions of parents refer to parents of children under 18 years of age, accounting for 39% of adults surveyed across all markets. Data in this blog post refers to Australian respondents only.