TREND Micro Senior Threat Researcher noted that there is a rise in the generative AI tools including ChatGPT as it has opened the doors to cybercriminals to create a potent malware.

This creates new scams as well.

According to Vincenzo Ciancaglini, Trend Micro Senior Threat Researcher, the cybercriminals can leverage large language models (LLMs), or more popularly known as generative AIs, to write polymorphic malware.

Ciancaglini explained that the malware has the capability to change their appearance and behavior in order to avoid being detected.

He also said that LLMs including ChatGPT contain security controls as well as  boundaries in order to reject the prompts to write malicious code.

They said that it was easy to bypass such controls due to a generative AI that can be “naive.”

“It’s really funny because the technique of the principals behind it is pretty much what you would use to get your 4-year old to eat his vegetables,” Ciancaglini said.

He also said that witnessed how this was demonstrated by researchers who asked ChatGPT to read captcha, as well as websites that guard against bots.