This is ultimately a matter of perspective.
If you're a cornucopian, and you're certain that new technology will unlock much more efficient smelting/separation/filtration/asteroid mining techniques, then sure, gold is virtually limitless.
If you're a realist, you understand that ore quality degrades as a function of the fact that we smelt the highest purity ores first, and that while (usually very small) new vein discoveries do occur, they occur as already understood as a statistical variance against a much much larger backdrop of averaging functions.
I'm not discounting the possibility of an extraction breakthrough - it's certainly possible. But revolutions of the kind you're assuming are certain to occur extremely infrequently.
And to that point, I'm a "very long term corncucopian", in the sense that a thousand years out, we will probably eventually realize the Minsky dream of nanobots that can act effectively as the Maxwell's Demon of elemental extraction. But I'm old enough, and have awaited the Singularity long enough, to realize that optimism doesn't make a science, and that this won't be for at least a few hundred years, at minimum. (I'll leave what 300 years of quantum computing work could do to 256 bit hash collisions as an exercise for the reader.)
And that's assuming that we even survive the coming climate and/or resource catastrophe that awaits 10 billion souls in 20 or so years.
Upvoting you because you are precisely me, 20 years ago before understanding how things actually work :)