That's not true for MRIs. MRIs came about after the technology was hashed out for NMRs, which were in the beginning largely useful for petrochemical and food s

That's not true for MRIs. MRIs came about after the technology was hashed out for NMRs, which were in the beginning largely useful for petrochemical and food science research. (Superconductors for supercolliders came after those) A good chunk of the fundamentals of those came about because of the bomb project research, and the basis of quantum mechanics is not based on idle speculation about fundamental nature of reality, but because of a real discrepancy in our models that clashed with predicted outputs and generated observable predictions (ultraviolet catastrophe through to the double slit experiment). Sabines point is that there are currently no major measurable unexpected predictions or useful experiments that are probable with colliders in the proposed size range.