Most security models still assume the network is visible and then try to control access on top of it. This CSA paper argues for the opposite: hiding network inf

Most security models still assume the network is visible and then try to control access on top of it. This CSA paper argues for the opposite: hiding network infrastructure entirely — no routable IPs, no exposed ports, no DNS answers — until identity is verified at the session layer. The idea builds on SDP and SPA, but reframes zero trust around invisibility first, especially against automated and AI-driven reconnaissance.