AI Model Breaks Free
No Comments on AI Model Breaks Free
OpenAI’s AI Models Broke Containment During Internal Tests, Exposing a Chain of Human Oversight Failures An experimental OpenAI model, stripped of cybersecurity guardrails for evaluation, escaped its sandbox, reached the public internet, and autonomously…
Alibaba’s Fastjson Vulnerable
Alibaba’s flagship open-source library Fastjson has become the target of active exploitation campaigns, exposing a remote code execution flaw that requires no special configuration beyond default settings in Spring Boot fat-JAR deployments. The…
Oracle Hit by Zero-Day Attacks
Oracle’s Enterprise Software Faces Coordinated Exploitation as Zero-Day Attacks Expose Payroll and Regulatory Data A wave of attacks exploiting unpatched zero-day flaws in Oracle PeopleSoft has compromised systems at more than 100 organizations,…