The gram is indeed defined by SI to be 1/1000th of a kilogram, it's not a random word I made up.
There is no logical difference between the definition you gave, and an alternative definition that says "the gram is defined by setting the Planck constant h to 6.62607015×10−31 J⋅s (J = g⋅m2⋅s−2), given the definitions of the metre and the second", and then defining 1 kg = 1000 g. Which is why I'm asking what this distinction actually means, if anything.
It meant something that feels more real when the kilogram was defined as the mass of a physical reference kilogram object in a vault in Paris, but that changed a few years ago.
There is no logical difference between the definition you gave, and an alternative definition that says "the gram is defined by setting the Planck constant h to 6.62607015×10−31 J⋅s (J = g⋅m2⋅s−2), given the definitions of the metre and the second", and then defining 1 kg = 1000 g. Which is why I'm asking what this distinction actually means, if anything.
It meant something that feels more real when the kilogram was defined as the mass of a physical reference kilogram object in a vault in Paris, but that changed a few years ago.