I think I just installed gpedit through dism. Just to point this out, I didn't do this to change the behaviour of the group-policy, but to try to find out if I had a local rule starting these scans. I expected to find one in either the registry or the group-policy set, but... couldn't find any. I half expected Windows update to have set new temporary rules, but didn't find that either. Meanwhile, you actually can successfully set these flags here, for when to allow the scans, and when to allow updates to install -- and they are respected by /most/ of the update runs(in the same way as the user-available settings that are hidden in fifty different places in Win11 now).
But not these specific script-runs. And that's more or less when I blew my top here. I had suspected before that there is a one-shot command-line trigger for the Windows defender scans coming in with the "security" updates. Because I keep seeing these updates being downloaded and triggered without any logs of what it is. But I thought it would have "su"-ed assumed the currently running user and launched the scan or whatever that way. Which then would have obeyed active hours, obeyed exceptions, and not freaked out when a program would be running and preventing it from being scanned, etc.
The thing is that this is not limited to the scans. I still get the situations when I have pending updates, for example, that then require a reboot. Where the system-account happily reboots my computer with no warning whatsoever during an "inactive" moment - that is, while I'm playing a game, or not doing something that specifically uses an active screen-context to display something. And it basically tells all running apps to close immediately to do the restart.
Meanwhile, the normal update packages, the ones that I can selectively install if I choose to - they are deferring to user profile settings, and will run happily once the computer is "free", or when you choose. They work just fine, and won't launch until you actually idle.
But the issue here comes with these critical hotfix pushes that I really can't see specifically. I've had one of these fume the laptop and then forcibly restart it in the middle of an exam (the exam program not being a full screen active context window). I've had a programming IDE open, along with some remote desktop contexts, that have the same issue just being wiped and suddenly restarted at US "night" hours. Imagine having something important going on and the computer just blanks and reboots, with no log of anything other than the device drivers panicking because their parent services are being forced to stop.
I know for a fact that several users from the same environment on work-provided laptops have the same thing, and that it clearly depended on whether or not the group-policy of your admin-group had, at some point, been added or not.
So the culprit here is, 99% certain, a script-command running something on default group-policies, from a fresh install, by the way, that are not "set". Where these script-admin people are expecting the work-environment to have been set, in spite of this not being the case for normal users.
And where most updates, and the normal deploy, obey the local group-policy rules (that Windows makes a huge issue out of you being able to set, by the way). But when the "work group"/system account's group policies are not set, or you haven't logged in with your school/work account at some point, you are running on another group policy setup (that I can't see on the home edition). And these defaults then introduce various issues when the local group policy (which is available to me, because it's part of my local "admin" control) is being ignored.
My issue here is that I can't document specifically what is being done, because I can't log the system account's activity. I can only see my user-account going haywire. And this is going to reoccur every month or so when these script-deploys are done.
Meanwhile, I'm sure there are thousands and thousands of users out there who are not aware of this, and think that Windows only breaking down for no reason just once a month is kind of ok. And that if the virus scan is toasting the computer for a month, that just means the computer is extra secure.
But this is kind of a big problem. Where system components can be changed via script, first of all, and invalidate any amount of driver-installs, and cause any amount of inconsistencies between what is installed and what is approved in the update streams. Or where virus-scans happily start changing settings back to .... well, pre-install-defaults without prompts, or add certain things that then cause issues with current user-settings. Or, when the virus-scans basically think that the running user-programs should be live-monitored with an exclusive, full attention monitor-process that drains system resources for days without stopping, while also failing to complete the scans, which then again perhaps launch exceptions, etc.
This stuff is insane. And the people writing these scripts have to know that default profiles cause these issues, surely. They have deployed them on thousands of computers, after all, not expecting them to ever be set.
You mentioned the "IT pro" community.. where this might be elevated, or I might get some help to find out what is going on exactly, and maybe a way to document this specifically, perhaps. I didn't get what you meant by that. Any links or anything would be great.