FBW could not change the laws of physics, if you use cambered airfoils, the lift of the wing has to be upward and that of the tail downward at 0° AOA to have some kind of positive stability, when the confusers go on leave.
Could I finally close the camber issue, after so much (debatable) discussion..
1) Mean camber would be -ve (e.g. on bottom surface, as installed)
2) Camber of a supercrital aerofoil section is not always that obvious, max thickness often far back (>50%) and strange camberlines used, but nevertheless, (1) applies.
3) Camber doesn't determine whether foil is lifting one way or other, it determines the ZLA (zero lift angle), Cl max, Cm and Cd curves.
4) Camber optimises the drag of the aerofoil for a given AoA... that is, in this case, the typical cruising Lift Coefficient would be determined and the shape (and hence camber-line) finalised, such that the low-drag-bucket was at that CL (-ve in this instance, however marginally - stability requires it)
5) The aerofoil is working in considerable (and varying) downwash from the mainplane, a stabilising situation.
6) Little of this has much to do with FBW or the somewhat aft CG during cruise; as has been said, this CG is set to offset cruise trim drag, not to create a zero or -ve stability FBW aircraft for maneouvrability reasons.
So the camber determines little, other than the setting angle of the THS and the actual Cd (drag) . As we know from watching amazing outside square loops during aerobatics, even quite + cambered foils such as on a Spitfire, can be flown upside down - they are just draggier and have lower maximum Cl (Clmax)