Do Son Microsoft's Project Perception is an agentic security system with red, blue, and green AI teams, reaching public preview on 3 August for the AI-attack era.
AI security
An AI-powered botnet was migrated in six minutes by a solo actor using Gemini CLI. TrendAI Research analyzed 200 session logs from "bandcampro".
Read our mid-July weekly threat intelligence report. Stay informed on active vulnerability updates, Fortinet flaws, and CISA KEV additions.
The Hugging Face breach was driven end-to-end by an autonomous AI agent attack. Internal datasets and service credentials were exposed.
Discover how new prompt injection attacks threaten AI security. Hackers use prompt injection techniques to manipulate AI agents and bypass safety controls.
Unit 42 warns phantom squatting attacks weaponize AI hallucinated domains, with 13,229 malicious URLs and 250,000 fake domains open to registration.
Three NVIDIA NeMo vulnerabilities (CVE-2026-24155, CVE-2026-24252, CVE-2026-24228) enable code execution. Update NeMo Framework to v2.7.3 now.
NVIDIA patches a critical 9.8 CVSS authentication bypass (CVE-2026-24207) in Triton Inference Server. Update to r26.03 to secure AI workloads!
Apple and Google form a rare alliance to block an EU mandate. Discover why Apple is fighting to stop third-party AI from accessing Android’s core architecture.
Google’s 2026 Android roadmap turns AI into a "Live Threat Detection" sentinel. Explore Android 17’s new anti-theft biometrics and banking fraud intercepts.
Sysdig warns: PraisonAI (CVE-2026-44338) exploited in under 4 hours. Attackers bypassed auth to hijack agents and drain API quotas. Update to 4.6.34 now!
Critical Alert: CVE-2026-44211 (CVSS 9.6) in Cline AI allows malicious sites to hijack your terminal and steal data via WebSockets. Update your CLI tools now.
LayerX uncovers a critical flaw in Anthropic’s Claude extension. Learn how attackers can hijack AI agents to steal data and bypass security policies.
Protect your AI apps. Spring AI issues urgent patches for three critical vulnerabilities causing data destruction, memory poisoning, and user data leaks.
Zscaler ThreatLabz exposes DeepSeek-Claw, a malicious AI skill targeting the OpenClaw framework to deploy GhostLoader and Remcos RAT. Secure your agents now!
Trend Micro uncovers a phishing campaign abusing Kuse.ai and VEC to steal credentials via markdown notes. Don't trust a domain just because of its reputation.
Lightning framework v2.6.2 & 2.6.3 are malicious. The "TEAM PCP" worm steals cloud secrets and poisons repos. Downgrade to 2.6.1 and rotate keys immediately!
ReversingLabs uncovers PromptMink: A DPRK campaign using LLM-generated npm packages to trick AI coding agents into installing crypto-stealing malware.
NVIDIA patches a critical 8.6 CVSS flaw in NemoClaw. Prompt injection could leak sensitive host secrets. Upgrade to v0.0.18 to secure your AI agent sandboxes.
NVIDIA issued an urgent patch for a critical 9.8 CVSS flaw in the FLARE SDK. Stop unauthenticated RCE and data tampering. Upgrade to v2.7.2 immediately.
Langflow patches a critical 9.6 CVSS path traversal vulnerability (CVE-2026-42048). Learn how an unsafe bulk delete function put entire filesystems at risk.
Critical CVSS 10 update for Gemini CLI (@google/gemini-cli). Fixes headless RCE and --yolo prompt injection risks. Upgrade to v1.83.7 to secure your pipelines.