He ChatGPT it is a tool with which we have been familiar since the end of 2022. The Artificial Intelligence mechanism, developed by the people of OpenAI, works with the typical legal restrictions of Google and the Internet in general.
However, as curiosity is one of the main characteristics of experts dedicated to technology, a group of South Korean researchers have trained this same Artificial Intelligence but with information based on the Dark Web.
For those who are not familiar with this concept, the Dark Web is a part of the Internet that is not accessible through conventional search engines and is associated with illegal and clandestine activities.
Due to the hidden and unregulated nature of the Dark Web, its content can range from illegal and dangerous information to content that is simply private or censored in certain countries.
So, when we meet a ChatGPT under these guidelines, it is normal to think that everything could end badly. But the reality is that no. According to a review by hypertextualSouth Korean scientists have a positive purpose for this kind of initiative.
It’s called DarkBERT and in order to gain access to the data on the Dark Web, the researchers connected it to the Tor network. Once this step was completed, they began to build a strong base of information collected during their forays into the dark corners of the Internet.
With what intention? Well, to have access to the working methods with which cybercriminals move.
Hipertextual explains that this technology offers various applications within this field. For example, it has the ability to identify websites specialized in selling ransomware or sensitive data. In addition, it can keep a constant watch on countless forums on the Deep Web, allowing you to detect any illicit information sharing and stay on top of suspicious activity.