Fair enough, but perhaps you missed the point of my post, too
Some of the discussions in this thread have been concerned with the use of virtual instruments being problematic in the face of Content ID's (allegedly) more aggressive and granular detection algorithms. I was merely pointing out that the same issues that apply to virtual instruments also apply to real instruments - it's all just sound as far as Content ID is concerned. Therefore, all music is potentially affected, not just music that uses virtual instruments, and that's potentially a nightmare scenario for the music biz.
I don't have the time nor resources required to build my own instruments from scratch every time I want to record a piece of music. Everyone uses the same guitars, pianos, basses, drums, etc., and virtual instruments, to record music because they are, after all, musical instruments and that's their intended purpose: to make music. Every symphony orchestra pretty much sounds the same, and that's what most people want - composers aren't trying to make an orchestra sound like "their own", and they shouldn't have to.
If Content ID is identifying and flagging individual instrument sounds, then that's definitely an overreach and impractical, for all the reasons stated above.