GOOGLE UNCOVERS PROMPTFLUX MALWARE THAT USES GEMINI AI TO REWRITE ITS CODE HOURLY

Summary from the TLDR InfoSec newsletter

Google identified PROMPTFLUX, an experimental malware, using a hardcoded Gemini API key to query Google's LLM with prompts for code obfuscation and AV evasion. It was able to self-modify and rewrite its source code to evade detection. Believed to be in testing by a financially-motivated actor, the malware includes a "Thinking Robot' that logs AI responses, persists via Windows Startup, and attempts propagation through drives and network shares, though it currently has no real attack capabilities. Security teams should monitor LLM-assisted malware like other variants, implement API key rotation, and be aware of threat actors bypassing AI safety via social engineering.