Beware, these fake ChatGPTs are a growing threat

The quarterly report from Meta, the parent company of Facebook, traces a trend: hacking attempts using ChatGPT as bait are growing.

Observe trends on the net and you will know what levers criminals use to try to trap Internet users. Everyone is rushing into cryptocurrencies? So it is sure that pirates use it to concoct all sorts of traps. Today, the trend is towards generative artificial intelligence and it is logically towards it that criminals turn.

These threats are not new. In February, for example, we noted the existence of a fake site imitating ChatGPT, designed to steal the identifiers of Internet users. In December, there was even talk of a diversion of the tool to write credible scams, thanks to convincing turns of phrase. But it seems like things have gotten bigger lately.

Malware that sometimes includes ChatGPT functions

In a waypoint shared on May 3, Facebook’s parent company gave some insight into this redirection of part of the nefarious ecosystem. Since March, Meta has detected ” dozens of malware families posing as ChatGPT and similar tools to compromise accounts on the Internet. »

ChatGPT is mentioned because it is the poster child for generative artificial intelligence. Since its public launch at the end of November 2022, it has been at the heart of all conversations related to AI. But there are others, specialized or not in the text. On the image side, for example, we can mention DALL-E, Stable Diffusion or Midjourney.

To lull the public or trick security systems in app stores, some malware manages to include operational features of ChatGPT anyway. ” This was probably to avoid any suspicion from stores and users », notes Guy Rosen, director of security at Meta.

Fashion is in ChatGPT so hackers adapt. // Source : Focal Photo

Links pointing to sites hosting this malware have been blocked in the Facebook environment – ​​more than a thousand URLs have been neutralized. Meta also says it has alerted the hosting platforms to take the necessary steps. Clearly, delete the files in question as soon as possible. Otherwise, the links are always accessible.

« For example, we have seen malicious actors creating booby-trapped browser extensions, available from official online stores, that claim to offer ChatGPT-related tools “, he continues. The game is worth the effort for criminals, because ChatGPT-like services drain maximum attention and audience.

Hackers, malware creators, spammers” pay close attention to what’s trending at the moment. They pick up on hot issues and popular topics to grab people’s attention “. Today, it is AI that captures the imagination and enthusiasm. Tomorrow it will be something else. And criminal campaigns will adapt.

Subscribe to Numerama on Google News so you don’t miss any info!

Understand everything about experimenting with OpenAI, ChatGPT

Source: Numerama by

*The article has been translated based on the content of Numerama by If there is any problem regarding the content, copyright, please leave a report below the article. We will try to process as quickly as possible to protect the rights of the author. Thank you very much!

*We just want readers to access information more quickly and easily with other multilingual content, instead of information only available in a certain language.

*We always respect the copyright of the content of the author and always include the original link of the source article.If the author disagrees, just leave the report below the article, the article will be edited or deleted at the request of the author. Thanks very much! Best regards!