OpenAI alerts 100+ orgs that its ‘misaligned models’ attempted to break in – or worse
www.theregister.com/security/2026/10/02/openai-alerts-100-orgs-that-its-misaligned-models-attempted-to-break-in-or-worse/5300891
Mostly ‘routine research tasks,’ and ‘some involved government websites, which our models often use,’ AI giant tells The Reg
Un agente de IA autónomo hackea al organismo de ciberseguridad holandés DIVD — y lo hace de forma errática y sin que nadie lo guíe
wwwhatsnew.com/2026/10/03/agente-ia-autonomo-hackea-divd-holanda-zammad-zero-day-2026/