I'm sure many will tell me why this cannot be done, but one can just as easily resize an entirely loaded round as they can an empty case.
In fact, before the advent of taper crimp dies, many would use the sizing die to "resize" the case neck and press it firmly into the sides of the bullet...on handgun rounds such as the .45ACP and 9mm.
In fact, factories do run loaded rounds through "final dimensioning" dies to ensure the finished rounds will cycle!
Semi-auto rifles have very low chambering force and are intolerant of oversized rounds.
When I first starting forming and "building" .300 AAC ammo, I immediately discovered a major problem with reliable chambering and extraction of unfired rounds. The bolt would refuse to close and once closed a mallet or "mortar" was needed to yank the round back out! This was with the cases formed and sized "die-to-case holder." While WAITING for a case gauge to slowly make it's way to my door, I decided enough was enough and starting shaving my sizing die, .0005" at a time. Then I'd grab a LOADED round that refused to chamber and run it through the sizer to set the shoulder back, and try to chamber again. I repeated this process until I reached a point where the bolt would close on the loaded rounds and extract without having to mallet tap the breech open. At this point - for whatever reason, I knew my rounds were correct for the chamber. I subsequently FIRED all those loaded rounds from that same rifle!
Here's the "twist." When my Lyman chamber gauge arrived, the first thing I did was drop in some factory NEW ammo only to discover it dropped in BELOW the "undersized" ledge! WHAT? And yes, this ammo had already been cycled through with 100% reliability. So then I grabbed my home-built rounds made from converted 5.56 brass and multiply sized as "loaded rounds" through my modified sizer die...they TOO dropped in just below the undersized ledge!
I went back and tested all the home-built ammo that I had not "yet" ran through the modified sizer die and as expected it dropped in level with the "acceptable" ledge which means is SHOULD work but it is actually too long at the shoulder datum to fit!
I currently own 5 different .300 barrels and ALL seem to have "short chambers"....almost as if the manufacturing spec for the round is something different than the published spec.
So with all that said, my POINT is that I have ran hundreds of rounds through the sizer to SET the SHOULDER back and not one ever blew up, nor should anyone EXPECT it to do so! Unless the primer is impacted by something during the process, the action of "squeezing" the powdered and loaded case is not dangerous at all.
The ONLY issue one might encounter with running loaded rounds through a sizer die is that in squeezing the case neck, the bullet is also being squeezed and could become loosened in the neck. I have encountered this effect when trying to get proper neck tension on the 5.7x28, .400 Corbon, and .357 Sig, and even bullets with a cannelure will loosen. Of course all these cartridges have ultra short necks with minimal bullet support! (This is WHY I stopped futzing around with bottleneck pistol rounds a few years back). So bottleneck cases in general are not good candidates for "whole cartridge sizing" but hey, if it were that or start pulling bullets, I'd try THAT first!
The .300 AAC has NOT proven to have a problem with bullets becoming loose, which is why I can state with confidence one can use whole-cartridge sizing to correct a shoulder problem.