I'm not sure that's true, because one thing that's done as part of these tests is a bunch of messing around with repository configuration that's meant to ensure a specific package update is available. We make sure at least one update is available so that we can test no "update is available" notification appears when booted live (suppressing update notifications in the live environment is a requirement) and the update notification does appear in the installed system.

On Silverblue, all that repo manipulation might not work at all, because we're not using regular package repos, right? Also, presumably we want Silverblue to notify us when an actual update (i.e. a newer versioned ostree, I guess?) is available - but in order to test that we'd need a reliable way to fake that situation, just as we use a side repo with a dummy python3-kickstart package to ensure an update is available for a regular RPM-based install...