Because content creators have a right to produce audio how they want.

If it’s a video of someone speaking in a quiet voice it shouldn’t have to have an average volume level of a new Metallica video. Forcing people to use heavy dynamic range compression to get it there would not be good.