Is the B/C data slope in AMS-02 data actually telling us something about the diffusion coefficient slope?

The measurement of the boron to carbon flux ratio provides relevant insights on the propagation of cosmic rays in the Galaxy. There is a common misconception about the fact that high-energy B/C data from AMS-02 directly provides the slope of the diffusion coefficient, implying that additional effects at play (convection, reacceleration, and destruction) can be neglected. Using the USINE code and taking into account all the relevant processes, we show that this assumption is not supported: the B/C data slope is not constant, its rigidity dependence being shaped in particular by inelastic interactions up to TeV energies. Heavier species are expected to be more impacted than lighter ones, and this should reflect in AMS-02 data, for which a softening of the slope with Z is predicted.