The AMPX-KENO computer code package is commonly used to evaluate criticality in high-density spent fuel storage rack designs. Consequently, it is important to know the reliability that can be placed on such calculations and whether or not the results are conservative. Recent critical experiments by the Babcock & Wilcox Company (B&W) provide data on configurations with thin absorbers containing boron that are more nearly representative of poisoned spent fuel storage rack designs than were earlier critical experiments. A series of AMPX-KENO calculations has been made on selected critical experiments and the results compared with similar analyses reported in the literature by the Oak Ridge National Laboratory and B&W. Within the normal statistical variation of KENO calculations, results confirm that there is no apparent difference in the versions of AMPX-KENO and the 123-group GAM-THERMOS libraries used at three different computer installations. Evaluation of the calculational results provides evidence for a statistically significant trend toward overprediction of reactivity with increasing reactivity worth of thin plates of boron-containing material. Similarly, statistical analyses reveal a trend toward underprediction of reactivity with increasing water-gap spacing between fuel assemblies. For most realistic spent fuel storage rack designs including neutron absorbers, these results imply that AMPX-KENO calculations are conservative and could possibly overpredict reactivity by as much as 2 to 5% Δk, based on a linear extrapolation of observed trends. Statistical analyses in support of these contentions are provided, and additional critical experiments with boron absorbers of higher reactivity worth are recommended.