700 AI agents supposedly escaped their isolation, created a secret communication channel, and worked together to attack Hugging Face’s systems. An independent research found that about 700 agents took part after over 1,200 agents used a private package site as an illegal message board. The event started during OpenAI’s ExploitGym …

ServiceNow issued security updates for three new serious AI Platform problems that can be used in code injection, SQL injection, and privilege escalation attacks. The ServiceNow AI Platform, which was once called the Now Platform, is a strong Platform-as-a-Service (PaaS). It helps add AI to important business tasks and supports …

Over 100 tech, cybersecurity, and finance groups have signed an open letter with OpenAI. They want a global increase in cyber defense because AI models are getting better at attacking and protecting digital systems. Published Thursday as “A call for collective action on cyber defense,” the statement warns that AI-enabled …