First principles calculations in the NiCoCr medium-entropy alloy predict a negative stacking fault energy (SFE) at T=0 K, implying an infinite Shockley partial dissociation distance. Many experiments at room temperature (RT) show however a finite dissociation. This discrepancy has been suggested due to solute strengthening that prevents the partial separation. Here, atomistic simulations in a model NiCoCr alloy having a negative SFE show that solute strengthening can limit partial separation at T=0 K but, the solute-induced barriers are easily overcome at RT and time scales of only 1 ns. Under experimental conditions (time scales of hundreds of seconds and longer), solute pinning is therefore insufficient to limit dissociation. Finite partial separations are thus presumably due to a positive stacking fault free energy at RT or short-range-order effects. It is argued here that the former is more likely than the latter.