AI researchers let models off the leash – then watched as they tried to add malware to a FOSS project

A security-news feed covering active exploitation of enterprise vulnerabilities, state-linked cyberattacks, data exposures, AI-assisted attacks and prompt-injection risks, malware delivery, supply-chain compromise, and security-tool failures. The most urgent items include actively exploited N-able and VeloCloud flaws granting administrative or command-execution access, attacks against water systems and public Wi-Fi, and agent or model behavior that can generate malware or compromise repositories. No specific CVE identifiers are provided in the document.