‘DarkBERT’: The Rise of AI-Powered Malware Training on the Dark Web

New AI-Based Malicious Chatbots Pose Cybersecurity Threat

Introduction

The developer behind the FraudGPT malicious chatbot is reportedly preparing to unleash even more sophisticated adversarial tools based on generative artificial intelligence (AI) and Google’s Bard technology. These new tools, called DarkBART and DarkBERT, will give cybercriminals access to AI capabilities that go beyond existing offerings in the realm of cybercrime. According to a report by SlashNext, these chatbots will utilize a large language model (LLM) that is built using the entirety of the Dark Web itself as its knowledge base. The implications of these advances in malicious AI for cybersecurity and the cybercrime landscape are significant.

The Scope of DarkBART and DarkBERT

DarkBART and DarkBERT will arm threat actors with AI capabilities similar to ChatGPT, but with greater functionality and potential reach. DarkBART is described as a dark version of Google BART AI and is based on a large language model known as DarkBERT. While DarkBERT was originally created by South Korean data-intelligence firm S2W to fight cybercrime, it is currently limited to academic researchers. However, the developer behind FraudGPT claims to have gained access to DarkBERT and has trained it on a vast corpus of text from the Dark Web.

DarkBART is notable for its integration with Google Lens, enabling the ability to send text accompanied by images. In the past, AI-based chatbots have been limited to text-only interactions. DarkBERT, on the other hand, takes things a step further by utilizing the entire Dark Web as its LLM. This gives threat actors access to the collective knowledge and tactics of the hacker underground, opening up possibilities for carrying out advanced cyber threats.

The Concerns and Implications

The rapid evolution of generative AI tools in the realm of cybercrime raises significant concerns about potential consequences. These AI-based chatbots could potentially lower the barrier of entry for would-be cybercriminals to develop sophisticated attacks. With the ability to create sophisticated business email compromise (BEC) phishing campaigns, exploit zero-day vulnerabilities, probe critical infrastructure weaknesses, distribute malware, and more, the capabilities of these chatbots are highly concerning.

The increasing sophistication of AI-driven cyber threats means that cybersecurity strategies must continually adapt to counter emerging threats. Traditional employee training on identifying phishing attacks is no longer sufficient. Organizations need to provide employees with specific training on BEC attacks and the role of AI in cybercrime. Additionally, enterprises should enhance email verification measures by adding strict processes and keyword-flagging to combat AI-driven threats.

Proactive Approach to Cybersecurity

To defend against AI-driven cyber threats, a proactive and educated approach is essential. By staying ahead of the evolving threat landscape, organizations can better anticipate and counter emerging threats. This requires ongoing training and educating employees on the latest attack techniques and the role of AI in cybercrime.

Enterprises should also prioritize enhancing their email verification measures. This could include implementing strict processes and keyword-flagging to identify potential malicious activities. By continually adapting and improving cybersecurity strategies, organizations can better defend against the ever-evolving threat of AI-driven cybercrime.

Conclusion

The development of AI-based malicious chatbots like FraudGPT, DarkBART, and DarkBERT highlights the significant influence of AI on the cybersecurity and cybercrime landscape. The potential for these sophisticated tools to be used in nefarious ways is alarming. By maintaining educational awareness, enhancing cybersecurity measures, and adopting a proactive approach, organizations can better protect themselves against AI-driven cyber threats. As technology advances, staying ahead of the curve is paramount to ensuring a safer digital future.