Gargi - Intrucept

AI Lazed Attacks Targets 600+FortiGate Firewalls; Amazon; What Makes AI a Powerful tool for Hackers

AI Lazed Threat Actors Compromised 600 +Fortinet FortiGate firewalls in 55 countries

Microsoft 365 Copilot Defect Exposes AI Summarizes of Confidential Emails

Microsoft 365 Copilot Vulnerability Bypasses DLP Policies, Summarizes Confidential Emails; Bug Tracked CW1226324

Summary :

A recently disclosed issue in Microsoft 365 Copilot caused the AI assistant to summarize confidential emails despite sensitivity labels and Data Loss Prevention (DLP) policies being configured.

The bug, tracked under CW1226324, allowed Copilot’s “Work Tab” chat feature to process and summarize emails from Sent Items and Draft folders, even when those emails carried confidentiality labels designed to restrict automated access.

Microsoft findings

Microsoft’s investigation revealed a code-level defect as the root cause. The flaw allows Copilot to inadvertently pick up items stored in users’ Sent Items and Draft folders, bypassing the confidentiality labels applied to those messages.

Although Microsoft categorized the issue as an advisory with potentially limited scope, the incident raises significant concerns regarding AI governance, trust boundaries, and enterprise data protection controls.

As per CSN the flaw allows Copilot to inadvertently pick up items stored in users’ Sent Items and Draft folders, ignoring the confidentiality labels applied to those messages.

Vulnerability Details

The issue happened because of an internal coding mistake in Microsoft 365 Copilot’s Work Tab chat feature. Due to this error, Copilot was able to access emails stored in Sent and Draft folders, even if they were marked as confidential.

In normal conditions, sensitivity labels and DLP policies should block automated tools from processing such emails.

However, because of this flaw, Copilot treated those protected emails as regular content and created summaries from them until Microsoft began deploying a fix in February 2026.

Attack Flow

Step	Description
Configuration	Organization applies confidentiality labels and DLP policies to sensitive emails.
Storage	Emails are stored in Sent Items or Draft folders.
Trigger	User interacts with Copilot “Work Tab” Chat.
Processing	Due to the code bug, Copilot accesses labeled emails.
Exposure	Copilot generates summaries of confidential content, bypassing expected DLP enforcement.

Source:0din

Why It’s Effective

DLP Control Bypass: AI processing occurred despite policy enforcement.

Trust Boundary Violation: Copilot acted as a privileged internal processor without honoring classification restrictions.

Compliance Risk: Potential regulatory implications under GDPR, HIPAA, ISO 27001, and industry frameworks.

AI Governance Gap: Demonstrates that AI systems must be independently validated against traditional security controls.

Broader Implications

This issue shows that AI tools inside business software can sometimes ignore security rules, even when protection like DLP and sensitivity labels are properly set. It proves that AI systems can create new risk areas that traditional security controls may not fully cover.

As more companies use AI assistants in daily work, security teams must regularly test and monitor how AI handles sensitive data. AI should be treated like a powerful internal system that needs strict oversight, not just a simple productivity feature.

Remediation:

Microsoft has initiated a fixed rollout and is monitoring deployment progress. However, organizations should take proactive measures:

Validate that sensitivity labels are now properly enforced with Copilot.

Audit Copilot usage logs and AI interaction history.

Re-test DLP enforcement across Sent and Draft folders.

Update AI governance documentation and risk registers.

Conduct tabletop exercises covering AI-driven data exposure scenarios.

Conclusion:
This incident highlights that AI integrations can introduce unexpected security gaps, even in well-configured enterprise environments. Organizations cannot assume that existing security controls will automatically work the same way with AI-powered features.

As AI adoption increases, companies must strengthen AI governance, continuously validate security policies, and monitor AI behavior just like any other critical system. Proactive testing and oversight are essential to prevent future data exposure risks.

Bypassing DLP policies by AI aided assistants signals huge security gap which needs to be addressed at enterprise level as AI tool taking over enterprise security posture cannot be undermined.

References:

Microsoft says bug causes Copilot to summarize confidential emails

Threat Landscape Expands in Supply Chain Due to Ransomware Attacks

How Ransomware Supply Chain Attacks Works

Chrome Security Updates by Google Released For Actively Exploited Zero-Day 2026

Chrome update released to patch a zero-day vulnerability that has been exploited in the wild.

Microsoft Patch Tuesday Feb-2026, Critical Azure RCE & Windows EoP & 0-Days

Microsoft’s February 2026 Patch Tuesday

Critical Flaw Identified in Fortinet Product ‘FortiClientEMS’; Security Updates Released

Fortinet released security updates for CVE-2026-2164

Fortinet has recently addressed a critical security vulnerability, identified as CVE-2026-21643, in its FortiClientEMS product. This flaw is classified as a SQL injection vulnerability, enables unauthenticated remote attackers to execute arbitrary code or system commands on affected systems by sending specially crafted HTTP requests.

Fortinet has released security updates to address a critical flaw impacting FortiClientEMS that could lead to the execution of arbitrary code on susceptible systems.

Technical Details

With a CVSS v3.1 base score of 9.1, this vulnerability is considered critical and poses a significant risk to organizations relying on FortiClientEMS for endpoint management.

The flaws affect the following versions –

FortiClientEMS 7.2 (Not affected)
FortiClientEMS 7.4.4 (Upgrade to 7.4.5 or above)
FortiClientEMS 8.0 (Not affected)

The vulnerability, CVE-2026-21643, resides in the FortiClientEMS administrative web interface.

Reason for the flaw or vulnerability to appear is caused by improper neutralization of user-supplied input in SQL queries. The flaw allows an unauthenticated attacker to send specially crafted HTTP requests to the FortiClientEMS GUI.

This resulted in the execution of arbitrary SQL statements, leading to unauthorized access, data exfiltration, privilege escalation and remote code execution (RCE) on any primary system.

Remediation

Immediate patching is strongly recommended to prevent potential exploitation, as the vulnerability allows attackers to bypass authentication and gain full control over the targeted system.

In addition to patching, organizations should implement the following best practices to reduce exposure and detect potential exploitation attempts.
Administrators should review web server and application logs for unusual or unauthorized HTTP requests targeting the FortiClientEMS administrative interface.
Monitoring for unexpected creation of administrative accounts or the execution of system commands originating from the FortiClientEMS host can help identify compromise.
Restricting network access to the FortiClientEMS management interface to trusted IP addresses and enforcing strong authentication controls can further reduce the attack surface.

There is currently no evidence of exploitation in the wild but the flaw has been termed a high-priority issue for all organizations using the affected product version, reason the attack surface is vulnerable.

Fortinet has since acknowledged that the issue has been actively exploited by bad actors to create local admin accounts for persistence, make configuration changes granting VPN access to those accounts, and exfiltrate the firewall configurations.

Conclusion:

The vulnerability is not present in FortiClientEMS versions 7.2, 8.0, or FortiEMS Cloud. The issue has been resolved in FortiClientEMS version 7.4.5 and later.

In the past similar Fortinet SQL injection and remote code execution vulnerabilities were found in Fortinet products and was targeted by cybercriminals and state-sponsored actors for financial benefits.

Sources: FortiClientEMS CVE-2026-21643: Critical Unauthenticated SQL Injection Vulnerability Allows Remote Code Execution