Researcher Tricks ChatGPT Into Building Undetectable Steganography Malware

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Watchdog
    PCHF Bot
    • Apr 2016
    • 19187

    #1

    Researcher Tricks ChatGPT Into Building Undetectable Steganography Malware

    Using only ChatGPT prompts, a Forcepoint researcher convinced the AI to create malware for finding and exfiltrating specific documents, despite its directive to refuse malicious requests.

    Continue reading…
Working...