Reminded of the idea that all progress distills down to incremental advances in instrumentation that compound and then BOOM into a paradigmatic transformation. CRUDing literal matter is probably one such boom.
What changed at Oak Ridge is the missing organ. Vision plus reward plus a physical actuator that can feel an atom. Once you have that loop, scale is an engineering problem, not a philosophical one. More tips. Parallel arrays. Faster feedback. Better force models. Better feedstock. The same curve we watched in lithography, in protein folding, in language models: a painful demonstration, then a night when the cost per operation falls through the floor.