One youngster, with the brutal honesty of youth, cautioned: “If your child is only talking to AI chatbots, they probably don’t have many friends at school.”
These statements on AI use among young Kiwis emerged as part of the first study looking at Generative AI and young people in New Zealand.
The centrepiece of the study was a survey of 13 to 24-year-olds living in New Zealand – although more than 90% of those surveyed were younger than 18.
Artificial Intelligence Safety Forum founder Shannon Novak is a co-author of the study, Generative AI and Young People in Aotearoa New Zealand.
Shannon Novak, director of Artificial Intelligence Safety Forum (AISF) and co-author of the study, was surprised by the level of engagement with the survey – receiving 1202 responses, after having expected about 400.
Netsafe CEO Brent Carey said the research shows AI is already embedded in how young people learn, create, solve problems and increasingly seek support.
“The conversation has moved well beyond whether young people should be using AI. They already are. The challenge now is making sure the adults around them understand the technology well enough to help them use it safely, critically and confidently.
“Young people aren’t asking adults to have all the answers. They’re asking us to be informed enough to have meaningful conversations without fear, judgment or panic.”
The study found that many young people wouldn’t turn to adults for help. Barriers included fear of getting in trouble, embarrassment and the belief that adults wouldn’t understand the technology.
It concluded: “This combination of unclear rules, unmet needs and fear of judgment may collectively contribute to the overall low confidence in seeking adult support.”
Novak told the Herald that AI could be “life-saving” for young people. It was always available and could provide ungated access to critical information in areas such as sexual health, drug interactions and mental health.
Mental health professionals, overseas and in New Zealand, report an increasing number of their adolescent clients seek mental health support from AI.
But Novak said it’s a double-edged sword: “What is the quality of the information that’s given? The biggest thing is just no accountability for anyone anywhere. A lot of these things are published overseas so they’ve got developers overseas who can just wash their hands of whatever happens on the app.”
This risk is compounded when young people aren’t comfortable reaching out to adults. However, some of the young people surveyed appear to understand the risks, with one saying: “The attachment you develop to an LLM [large language module] as it agrees with everything you say to keep you attached. I think this is very dangerous.”
Novak said another area of concern was the increasing use of AI by students to harass and defame teachers.
Novak told the Herald this includes fabricated slander videos that mock teachers or falsely accuse them of being a predator and sexually explicit deepfakes, with the latter overlapping with the problem of deepfake nudes at school in general.
Novak said the “content is convincing enough that staff have mentioned students being fooled by deepfakes”.
“Schools in New Zealand are struggling because, from what I can gather, most teachers have had no AI training.
“There’s confusion about how the tech works, and how seriously to treat it. And the law isn’t clear.”
Novak said New Zealand’s response was lagging behind other countries. “It’s not proactive, it’s reactionary.
“Australia has a reasonably good model to look at with the eSafety Commission, which would be great if we had an equivalent here.”
Or in the words of another young Kiwi: “Youth using AI is inevitable and should not be stopped. What should be done is guide them down the uncertain path, not to coddle but to prepare them.”
Chris Knox is a scientist turned data-journalist who investigates the stories behind the numbers, and creates interactives for Herald readers to explore them.