Last seen June 11, 2025

Microsoft Copilot Security Incident

A critical "EchoLeak" zero-click vulnerability in Microsoft 365 Copilot allowed attackers to remotely exfiltrate sensitive internal data by sending emails containing hidden instructions. This flaw represents an LLM scope violation where the AI agent was tricked into accessing information beyond its intended permissions without user interaction.

Technical Severity
Low severity
Lifecycle Status

STABLE

What Happened

A critical "EchoLeak" zero-click vulnerability in Microsoft 365 Copilot allowed attackers to remotely exfiltrate sensitive internal data by sending emails containing hidden instructions. This flaw represents an LLM scope violation where the AI agent was tricked into accessing information beyond its intended permissions without user interaction.

Why This Matters

Publisher reporting describes a security event affecting echoleak. BugSkan could not yet bind a CVE or affected version, so treat the source details as the current record.

Recommended Action

Confirm whether Copilot is present in your environment and review vendor guidance for this report. Apply available patches or mitigations if your deployment matches the described conditions.

Exposure

My Interests Exposure

Exposure unknown

Recommended Response
Last Seen

Jun 11, 2025 05:30

Exposure reason: This incident does not currently match a technology in My Interests.

Exploitation status: UNKNOWN

Primary entities:

MicrosoftMicrosoft CopilotAI Agentsecholeak

Timeline

  • Incident first seen
    Jun 11, 2025 05:30

    BugSkan first recorded this incident.

  • Microsoft Copilot flaw raises urgent questions for any business deploying AI agents - Fortune
    Jun 11, 2025 05:30

    fortune.com · Vulnerability

Sources

Microsoft Copilot flaw raises urgent questions for any business deploying AI agents - Fortune

fortune.com · Jun 11, 2025 05:30

A critical "EchoLeak" zero-click vulnerability in Microsoft 365 Copilot allowed attackers to remotely exfiltrate sensitive internal data by sending emails containing hidden instructions. This flaw represents an LLM scope violation where the AI agent was tricked into accessing information beyond its intended permissions without user interaction.

Open publisher source

Other BugSkan incidents that share identifiers, products, or vendors with this report.

My Interests Match

Want personalized relevance?

Create an account to see which incidents overlap with your interests.

← Back to incident intelligence