Has Artificial Intelligence Increased the Complexity of Cybercriminals?

Despite the numerous benefits of artificial intelligence (AI), malicious actors appear to be leveraging AI technology to enhance the complexity of cyber attacks.

Undoubtedly, the development of AI has brought significant advantages to many businesses. Over the past decade, the widespread adoption of AI and machine learning has enabled numerous organizations to successfully incorporate these technologies into their business practices, streamlining processes and improving productivity.

Today, companies across almost every industry are seeking opportunities to apply AI in their operations. However, like any industry, the advancements in AI have also brought about a dark side of technology.

Cybersecurity firm BlueVoyant reports that the cybercrime landscape has undergone significant changes in recent months. The company's report states that attackers continue to employ increasingly sophisticated techniques to write malicious code and highly realistic phishing emails.

What trends are we seeing now?

BlueVoyant's investigation, based on the findings of its threat intelligence team, reveals trends emerging in the field of AI and cybercrime.

Sale of AI tool credentials

Login credentials for ChatGPT have become highly valuable. They are being sold on the dark web, just like passwords for other online services. Cybercriminals often use information-stealing software to pilfer login credentials and sensitive data from unprotected devices, as this software is easy to deploy, especially if users are using outdated operating systems or have disabled automatic protection. Many users register with their company email addresses on OpenAI, and access data associated with these addresses fetches a slightly higher price on the dark web compared to data linked to personal email addresses, due to its sensitivity.

WormGPT and other open-source tools

While ChatGPT is designed to prevent its use for illegal activities, there are, of course, AI tools that are not subject to such restrictions. WormGPT is one such example, developed to assist cybersecurity professionals in testing malware generated using the tool and enhancing their defense against threats.

Despite the warnings from the developers on the website that the tool should not be used for criminal purposes, the fact remains that it can be used for illegal activities. BlueVoyant's threat intelligence has observed the development of a variant of WormGPT that can be used for malicious purposes and is available through dark web subscriptions. This variant can generate code in multiple programming languages and steal cookies and other valuable data from unsuspecting user devices.

Furthermore, WormGPT can also serve as a tool to support phishing activities. Attackers can craft highly persuasive messages with sophisticated language and wording, making it more difficult to identify deceptive emails. Additionally, it can be used to identify legitimate services and then exploit them for illegal purposes, such as SMS messaging services for large-scale phishing campaigns.

Future trends to expect

BlueVoyant's threat intelligence indicates that as AI continues to advance, cybercriminals may leverage these new technologies to create enhanced AI malware that can autonomously steal sensitive data and evade antivirus software detection.

Furthermore, AI is expected to aid cybercriminals in document forgery, making document verification increasingly crucial. Advanced AI tools can easily forge documents, enabling them to bypass online filters and engage in illegal activities such as creating false bank accounts and money laundering.

Looking ahead, concerns about the application of AI across various industries rendering jobs obsolete are widespread. However, in the field of cybercrime, experts at BlueVoyant point out that this concern has not yet become a reality. On the contrary, there is a shortage of IT skills that has extended to the realm of cybercrime, resulting in a significant demand for talent well-trained in generative AI (GenAI), some of whom unfortunately apply their expertise to illegal activities.

As organizations worldwide continue to integrate AI tools into their daily business practices, the "dark side" of AI activities is also rapidly increasing. Security teams need to be prepared for upcoming AI-driven cyber threats. However, at the same time, these threats are also increasing the demand for cybersecurity professionals, ultimately benefiting the long-term business environment.