AI Worm

An AI worm is a self-replicating malicious program designed to spread across networked systems and AI models without human intervention. These threats can propagate through compromised AI training data, model repositories, or system vulnerabilities, potentially corrupting multiple AI systems in an organization simultaneously. For enterprise governance, AI worms represent a critical risk because they can evade traditional cybersecurity controls and compromise the integrity and availability of AI infrastructure at scale.

1 item

ResearchGlobal2026-07-29

Unpatched AI Worm in Microsoft Copilot for Word Can Self-Replicate Through Enterprise Documents After 144-Day Disclosure Window

Security researcher Hakon Maloy has demonstrated a self-propagating prompt injection attack against Microsoft Copilot for Word that enables attacker-controlled instructions hidden in externally sourced documents to alter the content of newly drafted documents and copy the malicious payload forward into subsequent files. The vulnerability remained exploitable against the latest available model after a 144-day coordinated disclosure with Microsoft's Security Response Center. Because no vendor patch is available, enterprise document-handling policies and untrusted-source controls are currently the only available mitigations.

Microsoft Copilot prompt injection AI worm AI supply chain risk agentic AI document integrity Microsoft