Read here. The IPCC climate models almost complete failure at climate prediction has become an embarrassing joke within the general science community as these money-eating simulation efforts starve other science projects of funds. Almost on a weekly basis there is new research revealing the climate model failure fiasco, which likely will remain the case for the foreseeable future, per a recent study.
Wan et al. analyzed the Atlantic tropical bias that exists in the major IPCC climate models that prevents the coupled models from accurately reproducing Atlantic equatorial sea surface temperatures. This failure will not be solved in the near future they determine, which precludes these models being able to "predict" abrupt climate change.
"The authors write that "the notorious tropical bias problem in climate simulations of global coupled general circulation models manifests itself particularly strongly in the tropical Atlantic,"... they state that "the climate bias problem is still so severe that one of the most basic features of the equatorial Atlantic Ocean -- the eastward shoaling thermocline -- cannot be reproduced by most of the IPCC assessment report models,...as they describe it, "show that the bias in the eastern equatorial Atlantic has a major effect on sea-surface temperature (SST) response to a rapid change in the Atlantic Meridional Overturning Circulation (AMOC)."...ultimate implication of Wan et al.'s findings is, in their words, that "in order to accurately simulate past abrupt climate changes and project future changes, the bias in climate models must be reduced." But if "little or no progress" on this problem has been made in the tropical Atlantic "over the past decades,"..." [Xiuquan Wana, Ping Changa, Charles S. Jacksonn, Link Jia, Mingkui Lia 2011: Deep Sea Research Part II: Topical Studies in Oceanography]