Updating the taxonomy of failure modes in agentic AI systems: What a year of red teaming taught us
A surge in real-world attacks against agentic AI systems is reshaping how we think about risk.
Actionable threat insights
Get timely insights into emerging vulnerabilities, first‑finder discoveries, and evolving cyberattacker behaviors. Explore deep research and real‑world Microsoft Defender scenarios that show how proactive detection and quick action help organizations prevent compromise.
Refine results
Topic
Threat intelligence
Products and services
Publish date
Secure and verify every identity with Microsoft Entra
Microsoft Entra expands beyond identity and access management with new product categories such as cloud infrastructure entitlement management (CIEM) and decentralized identity.
-
-
Securing CI/CD in an agentic world: Claude Code Github action case
Microsoft Threat Intelligence identified a prompt injection pathway in Claude Code GitHub Action that allowed access to workflow secrets under specific conditions. -
AI brands as bait: How threat actors are using the AI hype in social engineering
As threat actors operationalize AI to accelerate attacks, they are also leveraging the wider global interest around AI itself as a social engineering lure. -
Reconstructing AI activity in investigations
Learn how to investigate AI activity in Microsoft 365 Copilot and Azure AI services using a structured, telemetry-driven approach. Prevent threats with Microsoft Defender
The Microsoft Defender family offers comprehensive threat prevention, detection, and response capabilities for everyone—from individuals looking to protect their family to the world’s largest enterprises.
-
Crypto Clipper uses Tor and worm-like propagation for persistence and control
Microsoft Threat Intelligence analyzed a cryptocurrency clipper campaign that combines clipboard theft, wallet replacement, Tor-based communications, and worm-like propagation. -
From package to postinstall payload: Inside the Mastra npm supply chain compromise by Sapphire Sleet
A poisoned npm package infected 140+ projects with a hidden payload. -
AutoJack: How a single page can RCE the host running your AI agent
AutoJack is a novel exploit chain showing how a single malicious webpage can turn an AI browsing agent into a remote code execution vector on the host machine. -
Guarding AI memory
What happens when threat actors target what AI remembers? Microsoft breaks down the risks and the defenses. Retain Microsoft Security Experts
Microsoft Security Experts are now available to strengthen your team with managed security services. Learn how to defend against threats with security experts.
-
StealC and Amadey: Breaking down infostealers and the cybercrime services that deliver them
On June 24, 2026, Microsoft’s Digital Crimes Unit (DCU) facilitated the takedown, suspension, and blocking of domains that formed the backbone of the StealC and Amadey infrastructure. -
Photo ZIP campaign targeting hospitality industry delivers Node.js implant for persistent access
Microsoft Threat Intelligence identified an active multi-stage intrusion campaign targeting hospitality organizations in Europe and Asia. -
Chromium extension uses AI‑related branding to redirect browser search
A malicious Chromium-based extension that spoofs the AI-powered answer engine Perplexity AI redirects browser search traffic using MV3 APIs and intermediary infrastructure. -
Securing AI agents: When AI tools move from reading to acting
MCP tool poisoning turns trusted AI agents into a control plane for data loss.
