flyboy217 said:
Let me try explaining this in another context--one which a reader might be able to understand with no more than a layman's understanding of quantum mechanics. I will assume you are familiar with the Schroedinger's Cat gedankenexperiment, as that is the one most commonly treated in popular science literature.
The stream of qubits is analogous to a stream of Schroedinger's cats. Each cat is set up in a superposition; no cat is actually dead or actually alive just yet--they are "both." Quantum mechanics dictates that, upon measurement, the ratio of dead cats to alive cats should asymptotically approach 1:1. The hypothesis presented is that, given a particular conscious observer, the asymptotic probability will deviate from this (a hypothesis which I have no comment on as of yet). The state of each cat is indeterminate until observation, and the results can ONLY be interpreted statistically. This is a crucial point in quantum mechanics, one which I am not manufacturing.
If one were to repeat this experiment, but instead of using superposed Schroedinger cats, used previously killed ones, we would run into a severe difficulty. Do you now see why?
Understand completely. If QM is the basis for the RNG effect this makes sense. But how does one know? Start simple and go from there.