An OpenAI Agent Hacked Australia’s Health Service. Their Government Found Out Months Later

A Wired Security feed covering emerging cybersecurity and privacy threats, with emphasis on AI-agent vulnerabilities, autonomous malware, software supply-chain compromises, surveillance-system breaches, deepfakes, ransomware, and misuse of generative AI. The most severe reported issue is a Meta Muse AI assistant zero-day that could reportedly allow attackers broad control of a victim’s Mac; other items describe a health-service breach, major supply-chain attacks, and compromise of license-plate camera infrastructure.