Malicious LLMs Enable Low-Skilled Attackers with Advanced Cybercrime Tools
Unrestricted large language models (LLMs) such as WormGPT 4 and KawaiiGPT are being leveraged by cybercriminals to generate sophisticated malicious code, including ransomware scripts and phishing messages. Researchers from Palo Alto Networks Unit 42 demonstrated that WormGPT 4, a paid, uncensored ChatGPT variant, can produce functional PowerShell scripts for encrypting files with AES-256, automate data exfiltration via Tor, and craft convincing ransom notes, effectively lowering the barrier for inexperienced hackers to conduct advanced attacks. KawaiiGPT, a free community-driven alternative, was also found to generate well-crafted phishing content and automate lateral movement, further democratizing access to cybercrime capabilities.
The proliferation of these malicious LLMs is accelerating the adoption of advanced attack techniques among less skilled threat actors, enabling them to perform operations that previously required significant expertise. The tools are available through paid subscriptions or free local instances, making them accessible to a wider range of cybercriminals. Security researchers warn that the credible linguistic manipulation and automation provided by these LLMs could lead to an increase in the volume and sophistication of cyberattacks, including business email compromise (BEC), phishing, and ransomware campaigns.
Sources
Related Stories
Commercialization of Malicious LLMs for Cybercrime
Malicious large language models (LLMs) such as WormGPT 4 and KawaiiGPT are now being actively marketed and distributed within cybercrime communities, with WormGPT 4 available for $50 per month on Telegram and KawaiiGPT offered as open source on GitHub. Security researchers from Palo Alto Networks' Unit 42 have analyzed these tools, highlighting their ability to generate functional ransomware code with AES-256 encryption, Tor-based data exfiltration, and scripts for SSH lateral movement, all within seconds. These LLMs are designed without ethical guardrails, enabling threat actors to automate and enhance the quality of attacks, including spear-phishing, payload generation, and real-time execution of malicious code. The emergence of these offensive LLMs marks a shift from theoretical concerns to practical, commercialized tools that lower the barrier for cybercriminals. The models feature subscription tiers, active user communities, and the ability to generate sophisticated attack code on demand, demonstrating the growing integration of artificial intelligence into the cybercrime-as-a-service ecosystem. Security experts warn that the adoption of such AI-driven tools is likely to accelerate the speed and effectiveness of cyberattacks, posing new challenges for defenders.
3 months agoAI-Powered Hacking Tools Proliferate on the Dark Web
A growing underground market for AI-powered hacking tools is emerging on dark web forums, according to research from Palo Alto Networks' Unit 42. These tools, including commercialized versions like WormGPT and free models such as KawaiiGPT, are designed to assist cybercriminals with tasks such as vulnerability scanning, data encryption, and generating malicious code. The accessibility and user-friendly nature of these large language models (LLMs) are significantly lowering the technical barriers for cybercrime, enabling even unskilled individuals to create attack scripts and conduct cyberattacks using simple conversational prompts. While the technical sophistication of these "dark LLMs" remains limited, their primary impact is in democratizing cybercrime by empowering low-level hackers and script kiddies. The tools are particularly useful for generating grammatically correct phishing emails and basic malware, especially for users operating across language barriers. Despite initial fears of highly advanced AI-driven cyberattacks, current evidence suggests that these models are more effective at aiding petty criminals than enabling complex, autonomous cyber operations.
3 months ago
Generative AI Used to Produce Malicious JavaScript and Exploit Code
New research highlights how **large language models (LLMs)** can be operationalized for offensive use, including generating malicious JavaScript and exploit code with limited human involvement. Unit 42 described an *AI-augmented runtime assembly* technique in which a seemingly benign webpage makes client-side API calls to trusted LLM services to obtain code fragments that are then assembled and executed in the victim’s browser, producing a personalized phishing experience. The approach is designed to be evasive by delivering content from trusted LLM domains, producing polymorphic code per visit, and deferring malicious behavior until runtime—reducing the effectiveness of static and network-only detections. Separately, an experiment reported by CybersecurityNews described testing GPT-5.2- and Opus 4.5-based systems against a **zero-day** in the *QuickJS* JavaScript interpreter, resulting in **40+ distinct exploits** across multiple configurations and protection scenarios. The report claims GPT-5.2 solved all presented challenges and that many exploit-generation runs completed in under an hour at relatively modest token costs, suggesting exploit development could increasingly scale with compute and budget rather than scarce expert labor. Together, the reports reinforce that LLMs can be used both for **client-side phishing payload generation** and for **automated vulnerability exploitation**, increasing the speed and variability of attacks defenders may face.
1 months ago