Aim Labs discovered a zero-click AI vulnerability named “EchoLeak” in Microsoft 365 Copilot and reported several ways to exploit it to Microsoft’s MSRC team.
The new attack method called “LLM Scope Violation” has been identified, which could also impact other RAG-based chatbots and AI agents. This finding marks a significant breakthrough in understanding how attackers can exploit AI agents by using their internal mechanics.
By infosecbulletin
/ Sunday , August 23 2026
More than 9,300 AWS access keys that were made public from August 2022 to August 2026 are still active, says...
Read More
By infosecbulletin
/ Sunday , August 23 2026
US Bank is looking into LockBit's claims about a breach and stolen data. The ransomware group says they will share...
Read More
By infosecbulletin
/ Friday , August 21 2026
Central Asian government agencies have been attacked in a cyber spy operation that used a small but different range of...
Read More
By infosecbulletin
/ Friday , August 21 2026
T-Mobile’s cybersecurity team reportedly physically cut a network cable connecting compromised infrastructure to the outside world. According to Bloomberg, the move...
Read More
By infosecbulletin
/ Friday , August 21 2026
Splunk has issued security fixes for 17 weaknesses in different apps and add-ons, such as Splunk MCP Server, Splunk AI...
Read More
By infosecbulletin
/ Friday , August 21 2026
Security experts have shown that expired credit cards can still be used. A study from the University of Massachusetts Amherst,...
Read More
By infosecbulletin
/ Thursday , August 20 2026
CERT Polska has alerted that bad actors are actively exploiting a security flaw in Zimbra Collaboration Suite to execute code...
Read More
By infosecbulletin
/ Thursday , August 20 2026
Operation CameraSwarm compromised 14,500+ Dahua IP cameras mostly in Ukraine and Russia. The operation lasted for at least 35 days...
Read More
By infosecbulletin
/ Thursday , August 20 2026
The Cl0p ransomware group has listed over 40 organizations that they say they targeted in a recent attack. This attack...
Read More
By infosecbulletin
/ Thursday , August 20 2026
Oracle has put out 943 new security updates in its August 2026 Critical Security Patch Update. These updates fix problems...
Read More

By using chains, attackers can automatically and unnoticed extract sensitive information from M365 Copilot, avoiding reliance on user behavior. The result is achieved despite M365 Copilot’s interface being open only to organization employees.
To successfully perform an attack, an adversary simply needs to send an email to the victim without any restriction on the sender’s email.
As a zero-click AI vulnerability, EchoLeak opens up extensive opportunities for data exfiltration and extortion attacks for motivated threat actors. In an ever evolving agentic world, it showcases the potential risks that are inherent in the design of agents and chatbots.
Aim Labs continues in its research activities to identify novel types of vulnerabilities associated with AI deployment and to develop guardrails that mitigate against such novel vulnerabilities.
Aim Labs is not aware of any customers being impacted to date. To read the full report click here.
40,000 + live internet cameras exposed globally !