The rise of AI agents is testing zero trust security, exposing risks that organisations can neither fully see nor reliably control ...