← Back to feed
AI SecurityEmerging1 sourceAug 27, 2024 · 00:30via Embrace The Red (AI agent security)

Microsoft Copilot: From Prompt Injection to Exfiltration of Personal Information

Brief

This post describes vulnerability in Microsoft 365 Copilot that allowed the theft of a user’s emails and other personal information. This vulnerability warrants a deep dive, because it combines a variety of novel attack techniques that are not even two years old.

I initially disclosed parts of this exploit to Microsoft in January, and then the full exploit chain in February 2024. A few days ago I got the okay from MSRC to disclose this report.

Read more on Embrace The Red (AI agent security)