#attack paths

Reduced safeguards in OpenAI's new GPT-5.6-Cyber model facilitate easier exploit development

Attackers are increasingly using exposed identities to navigate through complex network environments, moving from one domain to another via privilege escalation.