Learn about ChatGPT on the dark web

It is not denied that the world has entered the era of artificial intelligence forever, especially after the launch of models equipped with LLM, or “Large Language Model”, as in chatby OpenAI.

These super intelligent chatbots are able to interact as if they were human and can respond to almost anything.

This is possible because these AI systems have access to huge databases interconnected by artificial neural networks, in addition to being trained to communicate in the best possible way.

However, all of this advancement in artificial intelligence has alarmed business leaders and authorities around the world, who fear that these bots’ abilities could be used for evil in some way.

In this scope, DarkBERT has just been launched, a chatbot similar to ChatGPT, but trained using data obtained in the deep layers of the Internet, the so-called dark web.

According to its developers, who are South Koreans, DarkBERT is based on a data structure called RoBERTa, which pretty much performs the same function as its more well-known peers.

In short, to create a new chatbot, the researchers accessed dark web data networks in order to collect information that could be taught to DarkBERT.

With that said, DarkBERT now has its own database of information being transmitted on the dark web.

Once it became the talk of the internet, the new artificial intelligence began to be used as a tool that authorities could use to extract content from the dark web.

Notably, the dark web is a place primarily used by criminals to trade in prohibited items, such as stolen antiquities, contraband, and even human organs.

In addition, on the dark web it is also possible to track the activity of extremist groups, the spread of hate speech and currents harmful to support society as we know it.

And according to its developers, DarkBERT, as well as other super-intelligent chatbots, can still be improved and updated. To this end, the group will continue to update its database.