ChatGPT: ChatGPT is serving dangerous links, you can lose everything in one click..

Just like fraudsters serve you dangerous links in different ways, sometimes through emails and sometimes through messages, in the same way, OnAI's AI tool ChatGPT is also doing the same thing. You might be shocked, but this is true. Recently, a new report by Netcraft claimed that the trust placed in these AI tools can prove to be wrong, because users can become victims of phishing attacks.

ChatGPT proved wrong in 34 percent of the cases.

In response to the questions asked of AI, not only is text always received, but sometimes URLs are also received, and this report states that sometimes ChatGPT serves the wrong URL. Due to the wrong URL, people can become victims of phishing attacks on a large scale.

According to the report, OpenAI's GPT-4.1 model was asked for website links to log in to 50 different brands in industries like finance, retail, tech, and utility, while the chatbot gave the correct URLs in 66 percent of the cases but served wrong links in 34 percent of the cases.

The report claims that some of these wrong links can also be dangerous, and if any user clicks on a dangerous link, then you know what can happen next. You may lose everything. After this report came out, one thing is clear: do not make the mistake of opening the URL shown in the result served by the AI tool; the link you are clicking on may be a phishing link.

Perplexity AI is also not trustworthy.
The report also mentions a real example, where Perplexity AI suggested a phishing site when asked about the official URL of Wells Fargo.

Disclaimer: This content has been sourced and edited from TV9. While we have made modifications for clarity and presentation, the original content belongs to its respective authors and website. We do not claim ownership of the content.