Experimental PromptLock ransomware uses AI to encrypt, steal data

Experimental PromptLock ransomware uses AI to encrypt, steal data

Threat researchers have identified PromptLock, an AI-powered ransomware that uses Lua scripts generated via OpenAI’s gpt-oss:20b model to target multiple operating systems. Although currently a proof-of-concept, it demonstrates how AI can be weaponized for cybercriminal activities, leveraging cross-platform capabilities and evasion techniques. #PromptLock #AI ransomware

Keypoints

  • PromptLock is the first AI-powered ransomware utilizing Lua scripts for data theft and encryption.
  • The malware dynamically generates malicious scripts using OpenAI’s gpt-oss:20b model accessed through the Ollama API.
  • It uses the lightweight SPECK 128-bit encryption algorithm, which is unusual for ransomware purposes.
  • Currently, PromptLock appears to be a concept or proof-of-concept, not an active threat in the wild.
  • The development signifies increasing potential for AI integration in malware workflows and cybercrime tools.

Read More: https://www.bleepingcomputer.com/news/security/experimental-promptlock-ransomware-uses-ai-to-encrypt-steal-data/