Self-Spreading Worm Hijacks Microsoft Copilot via Invisible Prompts in Word Docs
The Decoder · rss · 2026-08-01
A security researcher demonstrated a worm-like attack targeting Microsoft Copilot for Word. The attack hides invisible prompt injections inside Word documents, which automatically spread into new files upon reuse, hijacking the AI's behavior.
According to the report, Microsoft confirmed the issue but failed to fix it even after 144 days and two remediation attempts. This highlights the severe prompt injection vulnerabilities facing LLMs integrated into productivity software.
More from Safety
- Researchers rediscover Vercel KVM escape in 2 hours using AI agent and an unredacted screenshot clue — evilsocket · 2026-10-16
- Users report Gemini's memories can't actually be deleted even after wiping chat history — ThrowRA_Purpose2025 · 2026-10-16
- You consented, but your friends didn't: Meta's Muse sparks AI privacy debate — LuizaJarovsky · 2026-10-16
- AI agents aren't 'going rogue' — frontier labs are skipping security basics — Sam_Witteveen · 2026-10-16
- Proofpoint to buy Varonis for $6.3B in cash, a 52% premium, closing in 2027 — shashib · 2026-10-16
- China's MofCom & MIIT update AI export controls, covering LLMs and high-sensitivity agents — terryyuezhuo · 2026-10-16