Even humans can flip a coin in such a way that the same side comes up in all branches of the wave function, as described by E.T. Jaynes, but IIRC he himself refers to that as "cheating".
I'm not sure that's what they mean either. I take them as saying 'humans can flip in a quantum-influenced way', not as 'all coin flips are quantum random' (as shminux assumed, hence the coin-flipping machine would be a disproof) or 'all human coin flips are quantum random' (as you assume, in which case magicians' control of coin flips would be a disproof).
If it's worth saying, but not worth its own post, even in Discussion, it goes here.