Detecting and analyzing prompt abuse in AI tools
Hidden instructions in content can subtly bias AI, and our scenario shows how prompt injection works, highlighting the need for oversight and a structured response playbook.
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
Products and services
Publish date
-
-
Contagious Interview: Malware delivered through fake developer job interviews
The Contagious Interview campaign weaponizes job recruitment to target developers. -
Malicious AI Assistant Extensions Harvest LLM Chat Histories
Malicious AI browser extensions collected LLM chat histories and browsing data from platforms such as ChatGPT and DeepSeek. -
Signed malware impersonating workplace apps deploys RMM backdoors
Signed malware backed by a stolen EV certificate deployed legitimate RMM tools to gain persistent access inside enterprise environments. 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.
-
OAuth redirection abuse enables phishing and malware delivery
OAuth redirection is being repurposed as a phishing delivery path. -
Threat modeling AI applications
AI threat modeling helps teams identify misuse, emergent risk, and failure modes in probabilistic and agentic AI systems. -
Developer-targeting campaign using malicious Next.js repositories
A developer-targeting campaign leveraged malicious Next. -
Running OpenClaw safely: identity, isolation, and runtime risk
Self-hosted agents execute code with durable credentials and process untrusted input. Modernize your Security Operations Center with Microsoft Sentinel
Microsoft Sentinel is a cloud-native SIEM solution powered by AI and automation that delivers intelligent security analytics across your entire enterprise.
-
Detecting and mitigating common agent misconfigurations
Agents are increasingly powerful. With that power comes risk: small misconfigurations, over‑broad sharing, unauthenticated access, and weak orchestration controls can create real exposure. -
Manipulating AI memory for profit: The rise of AI Recommendation Poisoning
That helpful “Summarize with AI” button? It might be secretly manipulating what your AI recommends. -
A one-prompt attack that breaks LLM safety alignment
As LLMs and diffusion models power more applications, their safety alignment becomes critical. -
Analysis of active exploitation of SolarWinds Web Help Desk
We are seeing exploitation of SolarWinds Web Help Desk via CVE‑2025‑40551 and CVE‑2025‑40536 that can lead to domain compromise; here is how to patch, hunt, and mitigate now.
