TechFlow, Nov. 7 — According to Decrypt, a recent report by Google's Threat Intelligence Group revealed at least five new types of malware leveraging large language models (LLMs) to dynamically generate and conceal malicious code. The North Korea-linked hacking group UNC1069 was found using Gemini to probe wallet data and create phishing scripts aimed at stealing digital assets. These malwares employ "real-time code generation" techniques, calling external AI models such as Gemini or Qwen2.5-Coder to evade traditional security detection. Google stated it has disabled the related accounts and strengthened security measures for model access.
Add to Favorites
Share to Social Media
