Some Alkali-GIC puzzling features, which may be of more general interest (specially to fullerides and conducting polymers) are discussed, namely : the validity of charge transfer heterogeneity deduced from NMR shifts; carbon-alkali hybridization in stage I viewed from 13-C shifts and relaxation rate; evidence of carbon sigma-pi hybridization in G.I.C.; possible long-range effects of inserted alkali atoms.