CI is really only significant on the programming side of things. Networking, systems administration, hardware, etc often don’t have that workflow.

sysadmins CI is hoping the machine still boots after an update. Unless you are lucky you got two identical machines to have a test env.

Also the case of 'learning that someone did updates directly on AWS console instead of terraform and losing 0.5 or more days cleaning up the resulting mess'

I see you have not discovered infrastructure-as-code (IaC). We have CI for our infra.

I’m familiar, the issue is IaC only works on a subset of system administration jobs.

Maintaining systems in a tunnel boring machine is a very different than an internet startup.