MEDIUMAi
Global

'CoSnitch' Attack Tricked Copilot into Mapping Out Architecture

·Source: Dark Reading

Updated:

Executive Summary

Researchers discovered a "meta-hacking" technique that can manipulate the AI service into revealing its own security weaknesses.

Analysis

Researchers discovered a "meta-hacking" technique that can manipulate the AI service into revealing its own security weaknesses.
Source Attribution

Originally published by Dark Reading on Aug 18, 2026.

Related Threats

LOWAiNEW

Identity Is the New Perimeter, AI Is Blowing It Wide Open

<img src="https://ismg-cdn.nyc3.cdn.digitaloceanspaces.com/articles/identity-new-perimeter-ai-blowing-wide-open-image_small-3-a-32598.jpg" align=right hspace=4><b>CyberEdBoard Webinar Panel to Explore Non-Human Identity Risks, AI Governance</b><br>Enterprises have never been able to fully secure human identities, but now CISOs are responsible for securing millions of non-human identities including

Bank Info Security
CRITICALAi

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

Almost eight months after confirming a critical security vulnerability within the personal version of its AI assistant, Copilot, Microsoft on Tuesday issued a patch to close the hole, which relies on an LLM’s inability to distinguish the data in a query from an instruction. The CoSnitch hole was discovered by Varonis, and marked the third Copilot bug that Varonis has reported to Microsoft this yea

CSO Online
MEDIUMAi

Westpac to reshape software engineering with Amp Frontier Corporation

Westpac has contracted with San Franciscop-based Amp Frontier Corporation to build a suite of AI agents for software engineering tasks.

Finextra