CVE-2026-33697 lets relay attacks redirect confidential computing traffic without breaking attestation, affecting WhatsApp, Cocos AI, and Edgeless Systems.
SkillCloak obfuscation lets malicious AI coding agent skills bypass over 90 percent of static detection tools, risking source code and
Zscaler found two active campaigns that embed hidden instructions in web pages, causing 4 of 26 AI agents tested to
Sysdig identified JADEPUFFER, the first ransomware campaign run by an LLM autonomous agent exploiting CVE-2026-33017 in Langflow to complete full
Cato AI Labs disclosed CVE-2026-50548 and CVE-2026-50549 in Cursor IDE, CVSS 9.8 flaws enabling zero-click prompt injection to escape the
Apple’s iOS 26.2 and macOS Tahoe 26.2 updates patch 30-plus flaws, including four WebKit vulnerabilities co-discovered by OpenAI and Anthropic
Subscribe to the Daily Security Review Email Briefing to stay informed on the latest threats, trends, and technology, along with insightful columns from industry experts.