Manuscript Oai265-A@2 · Oai265-A.tex · proposition · c0a34a45
Proposition: Uncalibrated records can conceal maximal changes of the tested object
Finite memory certification is not entropy calibration › Why zero simulated bias is not physical calibration · lines 1761–1776 · proof included or linked
Oai265-A@2#prop:fake
Without a trusted measurement model, even perfect three-setting records do not establish a Bell overlap or a memory-channel error.
Proof. An identity channel with ideal oriented Bell tests produces in every setting. A replacement channel with an apparatus that always reports , independently of its input, produces exactly the same attempted-run record laws. The channels have recovery behaviour separated by Lemma lem:constant, but their uncalibrated printed records coincide.