"Zero trust" sounds like a slogan, and on most marketing pages it is one. But there is a precise idea underneath, and you can read it off the patents. The old model trusted the network: once you were inside the corporate firewall, you could reach things. Zero trust throws that out. There is no inside. Every single request to reach a resource gets checked on its own — who is asking, from what device, in what state — as if it came from the open internet.
Zscaler's grant US12652286B2, "Systems and methods for directing and enforcing zero trust control on requests to destination services" (issued June 9, 2026; CPC H04L 63/10, access control), describes exactly that enforcement point: a request to reach a destination service is intercepted and a zero-trust policy is applied before it is allowed through. The trust decision happens per request, at the moment of access, not once at login.
Siemens' grant US12652309B2, "Method for providing real time zero trust security in a shared resource network" (issued June 9, 2026; CPC H04L 63/1433), pushes the same principle into an environment where many parties share infrastructure — the harder case, because there is no natural perimeter to even pretend to trust. Real-time evaluation is the point: trust is continuously re-decided, not granted once and assumed.
The practical takeaway for defenders: zero trust is not a product you install, it is an architecture you converge toward, and its hard part is the policy engine that decides every request. Two different assignees — a cloud-security pure-play and an industrial conglomerate — independently patenting per-request enforcement tells you this is the settled direction, not a fad.
Why it matters for breach stories: most damaging intrusions involve lateral movement — an attacker who gets one foothold and then roams a network that trusted its own insides. Zero trust is the architectural answer to lateral movement specifically. When a disclosure describes an attacker pivoting from one system to another, that is the failure mode these patents are built to deny. The grants are methods, not guarantees; but they show, concretely, what "never trust, always verify" looks like when an engineer has to actually build it.