The scramble to use ChatGPT for cybercrime has seen a seven-fold increase in dark web forums where hackers discuss how to manipulate the chatbot. That’s according to a new study by cybersecurity company NordVPN.
According to the research, the number of new posts on dark web forums about the AI tool rose from 120 in January to 870 in February — a 625 percent increase. Forum posts related to ChatGPT increased 145% from 37 to 91 in a month as discussions surrounding the bot's abuse became the hottest topic on the dark web.
ChatGPT is designed to produce malware and scripts
In early January, many threads focused on how malicious actors could trick ChatGPT into producing simple malware, a vulnerability that has since been fixed. A month later, the trend in the hacker community was towards more aggressive measures. The posts outline plans to take control of the chatbot and use it to wreak havoc. Dark Web forum titles (which are hidden from normal search engines and visited by anonymous hackers around the world) included “How to break ChatGPT”, “ChatGPT jailbreak 2.0, ChatGPT – progression of malware” and “ChatGPT as a phishing tools”.
With ChatGPT, criminals could use its artificial intelligence to stop scams – such as e.g. B. Love Scams – to be carried out on an industrial scale and to attack multiple victims at the same time. Once a hacker has taken over the chatbot, they can remove its security restrictions and use it to create malware and phishing emails, spread hate speech, or even spread propaganda.
Many new ways with ChatGPT
“ChatGPT has opened many doors to new worlds and is transforming the world of work, education and how we use information. Like any new technology, it can be leveraged for positive uses, such as detecting vulnerabilities in smart contracts.
At the same time, the dangers for users are not insignificant. This is also shown by NordVPN's investigation into the increase in forum posts on the dark web. This allows criminals to use AI-powered vulnerability detection for malicious purposes. This also includes the misuse of quickly finding and summarizing personal information in order to attack companies and people in a more targeted manner. Even more dangerous is bypassing security checks and filters in ChatGPT and other AI providers to trick the program into making racist or hateful statements,” said Marvin Petzolt, cybersecurity expert at Nord Security.
More at Nordsecurity.com
About North Security
Founded in 2012, Nord Security is home to advanced security solutions that carry the Nord brand. We build robust cybersecurity tools for personal and business use, including the world's fastest VPN service, NordVPN.