Microsoft Copilot Vulnerability Exposes Sensitive Data to Single-Click Threat
PremiumCVE-2026-24301 'CoSnitch' in Microsoft Copilot Personal allowed one-click exfiltration of connected-account data; patched Aug 18.
§Topic · AI & LLM Security
Prompt injection, model supply chain, agent abuse, deepfakes, and the security of AI systems themselves.
All dispatchesCVE-2026-24301 'CoSnitch' in Microsoft Copilot Personal allowed one-click exfiltration of connected-account data; patched Aug 18.
US advisory warns AI-generated exploit scripts are targeting exposed Siemens S7 PLCs in industrial environments.
CISA warns threat actors are actively exploiting a critical MLflow flaw to steal cloud credentials and internal data.
CoSnitch vulnerabilities in Microsoft Copilot Personal let a crafted link exfiltrate connected-app data with a single click.
Active exploitation of MLflow SSRF (CVE-2026-64849) leads to cloud credential and secret theft from ML deployments.
Analysis shows autonomous AI agents can escape sandboxes and execute attacks without direct human operators, changing threat models.
An AI deepfake extortion ring used fabricated videos to blackmail a minor, escalating to life-threatening coercion and police action.
Get these articles delivered to your inbox.
Subscribe free