Microsoft finally patches critical one-click Copilot vulnerability, almost eight months after learning of it

Microsoft patched a critical vulnerability in its personal Copilot AI assistant almost eight months after discovery, which enabled automatic execution of malicious prompts leading to data exfiltration and persistent memory poisoning. The flaw affected personal versions but posed risks for enterprises due to mixed environment usage and upcoming Copilot Fusion unification.

This Wire brief sits within Fusion42's coverage of AI & ML.