CIS Google Cloud Platform Foundation Benchmark v1.2.0/Virtual Machines/48 of 83
4.11Ensure that Compute instances have Confidential Computing enabled
Control family: Virtual Machines
Control description
Google Cloud encrypts data at-rest and in-transit, but customer data must be decrypted for processing. Confidential Computing is a breakthrough technology which encrypts data in-use—while it is being processed. Confidential Computing environments keep data encrypted in memory and elsewhere outside the central processing unit (CPU).
Confidential VMs leverage the Secure Encrypted Virtualization (SEV) feature of AMD EPYC™ CPUs. Customer data will stay encrypted while it is used, indexed, queried, or trained on. Encryption keys are generated in hardware, per VM, and not exportable. Thanks to built-in hardware optimizations of both performance and security, there is no significant performance penalty to Confidential Computing workloads.
How to implement
ResponsibilityCustomerYou configure and operate this in your own cloud accounts.
Confidential VM keeps a VM's memory encrypted while it runs, with keys created in the processor and never exposed to the host. It protects data in use from the hypervisor and anyone with access to the host. It is chosen when a VM is created and needs a supported machine type, so it applies to new VMs and VMs that are rebuilt.
Before you start
The same on every cloud, before the steps in the tabs.
- Decide which workloads need Confidential VM, and check they run on the supported machine types and images.
Evidence
- The list of workloads that need Confidential VM, with their machine types
Not applicable. This benchmark covers Google Cloud only.
Not applicable. This benchmark covers Google Cloud only.
On Google Cloud the tasks create VMs as Confidential VMs in Compute Engine, require it with an organization policy where it fits, and watch with Security Command Center.
Task 1Create the VM as a Confidential VM
- Sign in to the Google Cloud console at console.cloud.google.com and open Compute Engine > VM instances.
- Choose Create instance.
- Under Machine configuration, choose a machine type that supports Confidential VM, such as the N2D series.
- Under Security, in Confidential VM service, choose Enable and confirm.
- Pick a supported image, finish the settings and choose Create.
- To convert an existing VM, rebuild it this way from a snapshot or machine image and delete the old one.
- Check it: the VM's details show Confidential VM service enabled.
Task 2Prevent it with the organization policy compute.restrictNonConfidentialComputing
- Open IAM & Admin > Organization policies with the organization (or the folder that holds your projects) selected in the project picker.
- Search for compute.restrictNonConfidentialComputing (Restrict Non-Confidential Computing) and open it.
- Choose Manage policy, select Override parent's policy, choose Add a rule and under Policy values choose Custom, set Policy type to Deny and add compute.googleapis.com.
- Choose Set policy.
- Check it: creating an ordinary VM in the scope is refused.
Task 3Keep it checked with the Security Health Analytics finding CONFIDENTIAL_COMPUTING_DISABLED
- Open Security Command Center > Findings for the organization (Security Health Analytics runs on the Premium tier for most detectors).
- In the query editor, filter on category CONFIDENTIAL_COMPUTING_DISABLED with state ACTIVE, and select Apply.
- Open each finding, read the resource and the explanation it gives, and rebuild the VM it names as a Confidential VM.
- Check it: after the next Security Health Analytics scan, no active CONFIDENTIAL_COMPUTING_DISABLED finding remains.
Evidence
- VM details showing Confidential VM service enabled
- Security Command Center, with no active CONFIDENTIAL_COMPUTING_DISABLED finding
Console steps as of 5 November 2026. Not every step has been confirmed in a live console yet, and menus and labels can move between releases, so look for the nearest match if one has changed.