New Zealanders overconfident in spotting AI fakes

A recent report highlights that a significant portion of New Zealanders believe they can identify AI-generated fakes despite engaging in risky online practices. The study indicates that 66% of adults feel confident they would instantly recognize AI fakes, yet 40% have either shared or contemplated sharing sensitive data with an AI tool. This discrepancy between self-assurance and behavior is termed an “AI confidence trap” in Avast’s 2026 Safe Tech Report, which is based on a survey of 1,000 New Zealand adults. The research also revealed that 21% of respondents have shared or considered sharing a CV with an AI chatbot, while 17% have done the same with a work document.

The report emphasizes that risks extend beyond isolated interactions, as data entered into AI systems may persist post-conversation, potentially leading to exposure through breaches, unauthorized access, or training data usage. Richard Watts, Avast’s head, stressed that the issue isn’t merely about detecting synthetic content. “Being confident you’d spot an AI fake and knowing when and how to protect your data are two different skills – but confidence in one can convince people they’re safe with the other too, and that’s the trap,” Watts said. “Chatbots aren’t confidants.

Check what a tool actually does with your data before you share anything, and remember it’s still somewhere else your information could be exposed. If you wouldn’t want it repeated to a stranger, don’t hand it over to the bots,” he added.

Personal Data Habits

The research found that many consumers recognize common warning signs of fraud but still dismiss them in the moment. Among the signals respondents were most likely to ignore were an offer that looks too good to be true, cited by 42%; a request for personal details to “verify” identity, cited by 39%; and pressure to act quickly, cited by 38%. Avast linked those findings to wider scam patterns identified by Gen Threat Labs, saying AI-generated fraud often depends less on technical sophistication than on familiar tactics designed to trigger urgency, trust, fear, or greed.

Scam Patterns

Gen’s recent threat data captured tens of millions of AI-related video interactions globally during the first half of 2026, with approximately one in 30 identified as fraudulent. This implies that even a minor proportion of malicious content can result in a substantial influx of persuasive fake material. Michal Salat, Gen Threat Labs’ threat intelligence director, stated that scam patterns have remained consistent despite evolving tools. “What we see in the threat environment reinforces what consumers are telling us here.

Spotting that something was made with AI is only part of the challenge,” Salat said. “Through our Fearless Planet Index, we see how scams are evolving in the real world, and AI is increasingly making the first impression more convincing. But underneath, attackers are still pulling the same psychological levers – urgency, trust, fear, and the promise of something too good to pass up. The technology may be changing, but the human behaviours scammers exploit remain remarkably consistent,” he said.

Consequences of Scams

The survey revealed that numerous New Zealanders handle suspicious activity discreetly instead of reporting it. When faced with an odd message, call, or direct contact, 43% would delete it without informing anyone, and 20% would probe it independently. The data also highlights real repercussions. Avast noted that 51% of NZ scam victims experienced financial losses.

The threat is now broadly acknowledged, with 24% of NZ adults admitting a listed scam red flag could realistically ensnare them. The survey, conducted online by Dynata and adjusted by age, gender, and region to mirror the national demographic, shows this awareness. Practical implications include heightened vigilance regarding AI-generated content risks and proactive measures to safeguard personal data, such as exercising caution when sharing confidential information with AI tools and validating the legitimacy of messages and offers.

Elderly relatives pose a significant concern. Of those surveyed who monitor older family members with risky online behaviors, 72% express worry about them clicking dubious links. Within this group, 14% reported a relative falling prey to phishing scams, whereas 54% credited their guidance in averting such fraud.