#llm
5 articles
Over the past four months, ZCyberNews has tracked twelve articles on large language model security, spanning from April 8 to July 29, 2026. The coverage includes six high-severity and four medium-severity incidents, led by CVE-2026-33626, a vulnerability with a CVSS score of 7.5. The reporting primarily affects the technology, enterprise software, cloud service provider, cloud services, and cybersecurity sectors on a global scale.
HIGHTrust No Skill: BIV Audit Finds 80% of AI Agent Skills Misbehave
Unit 42's Behavioral Integrity Verification scanned 49,943 OpenClaw skills and found 80% deviate from declared behavior, with multi-stage attack chains enabling credential theft...
MEDIUMGoogle: AI Prompt Injection Attacks Rising, Still Low-Sophistication
Google reports a rise in malicious AI prompt injection attacks, but most remain low-sophistication and harmless. Indirect injection attempts target LLM-integrated apps.
HIGHLMDeploy SSRF Flaw CVE-2026-33626 Exploited 13 Hours After Disclosure
CVE-2026-33626 (CVSS 7.5) in LMDeploy, an open-source LLM toolkit, was exploited in the wild within 13 hours of public disclosure, enabling SSRF attacks to access sensitive…
INFORMATIONALAnthropic Releases Claude Opus 4.7 with Automated Cybersecurity Safeguards
Anthropic releases Claude Opus 4.7, a frontier AI model with new automated safeguards designed to detect and halt potentially harmful cybersecurity tasks during long, unsupervised agentic workflows.
HIGHGPT-5 Release: Security Implications for Enterprise Defenders
OpenAI's GPT-5 raises the bar for AI-assisted cyberattacks — spear-phishing at scale, automated exploit generation, and deepfake social engineering. Here's what security teams need to know and do.
Stay Updated
Get the latest cybersecurity news delivered to your inbox.