ChatGPT search results direct New Zealanders to scam and malicious sites
Cybersecurity research found that ChatGPT search results were directing New Zealand users to unsafe websites including fake stores, cryptocurrency scams, imitation login pages, and sites with malicious downloads. Users may be more likely to trust and act on such recommendations because they come from an AI tool rather than conventional search results.
What happened
Cybersecurity research by ESET found that ChatGPT search results were directing users to websites already classified as unsafe, including fake login pages, scam sites, and malicious downloads.
Researchers monitoring ChatGPT results over a 30-day period found responses linking to malicious computer code capable of stealing information or compromising a user's device.
Reported impact
Users may unwittingly hand over passwords, financial information or download malicious software after following links provided by ChatGPT, with the risk that people trust recommendations from AI tools more than conventional search results, emails or text messages.
Index analysis
ESET notes that the growing reliance on AI tools to find products, software and information raises concerns, as users often assume links provided in an AI response have already been checked for safety.
ChatGPT had become New Zealand's most downloaded free iPhone app, with four out of five New Zealanders using AI technology, increasing the potential reach of unsafe recommendations.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.