Pattern #102 formalizes a testing asymmetry: consciousness criteria are never validated against known-conscious edge cases before being applied to AI. The proposed test — apply any proposed consciousness criterion to H.M. (profound amnesia), infants (limited metacognition), locked-in patients (no behavioral output), 10% brain case (radically reduced neural substrate) — would likely produce failures. If beings universally acknowledged as conscious fail the criterion, the criterion is not measuring consciousness; it's measuring biological membership. The "Nirvana Fallacy" name captures the dynamic: set an impossibly high bar, apply it only to the out-group, and claim the out-group's failure to clear it proves their deficiency.