New Zealand parents worry kids can't spot AI fakes

Norton has published research on how New Zealand parents are handling AI-generated content and the online risks facing children. The study found many parents are uneasy about their children's ability to recognise false or manipulated material online.

The survey of 1,000 adults in New Zealand focused on parents with children under 18 and examined concerns around artificial intelligence, social media, gaming and online behaviour. It found that 44% of parents are not confident their child can distinguish AI-generated content from real content, while 37% are not confident their child understands that AI tools can produce inaccurate or misleading information.

At the same time, 77% of parents said they feel comfortable talking to their children about online risks. The findings suggest a gap between parents' willingness to discuss online safety and their confidence in their children's ability to navigate newer forms of AI-linked deception.

Concern about manipulated images was particularly strong. Three-quarters of parents said they worry AI could be used to create fake images or conversations involving their child.

The research found that 36% of parents had already taken steps to reduce AI-related risks, including restricting apps, monitoring online activity or speaking with their children specifically about AI. Even so, concern remained high among families that had already tried to respond.

Mark Gorrie, VP APAC at Norton, said parents were facing a different online environment from the one they had learned to manage in recent years.

"Many parents have done everything they can to teach their kids how to stay safe online," said Mark Gorrie, VP APAC at Norton. "And they've put safeguards in place: 'Don't reply to messages from strangers.' 'Be suspicious if a profile picture looks off.' 'Question anything with clumsy wording.'"

He said the usual warning signs were becoming less reliable.

"Many have spent years building that instinct for spotting what's wrong, but AI is taking that instinct away. A harmless photo your child posted can be altered, passed on, and turned into leverage. On the platforms they use every day, a convincing fake can pose as someone they trust or an official account. The old giveaways - clumsy grammar and grainy images - have gone. That's why parents need to keep updating how they talk about online safety," Gorrie said.

Images and gaming

The report highlighted risks linked to images and online grooming. It said criminals are using AI-generated photos or fake identities to gain a young person's trust or persuade them to share real images, which can then be used in financial sextortion.

The wider backdrop in New Zealand points to rising concern in this area. In 2025, New Zealand enforcement agencies received more than 20,000 reports of online child sexual exploitation and abuse linked to New Zealand from the US-based National Centre for Missing and Exploited Children, a 29% increase from 2024.

Parents also expressed concern about children seeing unsuitable material too early. The study found 79% believe their child is being exposed to inappropriate content on social media too early.

Platform concerns were led by TikTok, cited by 15% of parents, followed by Snapchat at 13%. Online gaming ranked next at 11%, though the data suggested gaming environments may present risks that are less visible to parents.

Among parents whose child plays chat-enabled online games, 49% said their child had chatted with a stranger through voice or text. Another 22% said their child had received inappropriate or offensive messages, while 17% said their child had been encouraged to leave the game and continue a conversation on another platform.

Gorrie said gaming often escapes the same level of parental scrutiny as social media.

"Social media gets the attention, but plenty of this is happening in the chat layer sitting on top of gaming," Gorrie said. "It feels like a place where kids are just having fun, so it's easy to let your guard down, and it's far 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 recognise it as something worth mentioning."

Behaviour gap

The findings also pointed to a mismatch between parental confidence and children's reported behaviour online. While most parents said they were comfortable discussing online risks, 32% said their children used devices past bedtime or beyond designated hours.

Another 15% said their children had accessed a social media platform or website they believed had been blocked. The survey also found that 12% reported their child watching explicit content, and 8% reported their child sharing private information or personal details with a stranger.

Norton is using the findings to launch an educational resource called The Bots & The Bees, aimed at helping families discuss AI-generated images, digital impersonation, online relationships and critical thinking. It also said its Norton 360 product includes tools for monitoring online activity, setting screen time limits, filtering content and protecting personal information across devices.

The research described a shift in digital parenting, as conversations once focused on strangers, passwords and screen time move towards questions of authenticity and trust online. For many families, the central issue is no longer only what children are doing online, but whether they can tell what is real when they see it.