<p indent="0mm">The Cretaceous-Paleogene (K-Pg) mass extinction that ended the age of the dinosaurs occurred 66 million years ago, with competing hypotheses including the Deccan Traps volcanic eruption, climate change, and sea level change. However, it is the Chicxulub asteroid impact that has been widely accepted as the most plausible cause. Whatever the reason, the Earth’s ecological environment changed dramatically, leading to a mass extinction of both terrestrial and marine life. Primary producers lie at the base of the food chain and support higher trophic levels, so play an important role in the marine ecosystem. The predominant primary producers in the ocean, coccolithophores (phytoplankton with small CaCO<sub>3</sub> scales covering their cell surface), were among those most severely devastated by the K-Pg mass extinction. So, studying fossil coccolithophores across the K-Pg boundary can provide a better understanding of the destruction and post-event recovery processes of marine primary production (PP) and export production (EP), and the changes and controlling mechanisms of the marine biological carbon pump in the current global climate change. The four hypotheses previously proposed to interpret the post-event primary production, namely strangelove ocean (collapsed PP), living ocean (collapsed EP), resilient ocean (continued PP and EP), and heterogeneous ocean (geography-dependent responses of PP and EP), are illustrated and compared, and their pros and cons reviewed. A dataset is compiled from multiple locations in the global ocean basin including Deep Sea Drilling Program Site 465, Ocean Drilling Program Sites 690, 1210, and 1262, and the outcrop at El Kef (Tunisia), with each location having a high-precision cyclostratigraphic time control and an ocean surface-to-deep (bulk carbonate-benthic foraminifera) <italic>δ</italic><sup>13</sup>C vertical gradient. This global dataset is analyzed to explore the changes in the efficiency of global marine biological carbon pump after the K-Pg boundary event. Three important patterns are observed: (1) Prior to the K-Pg event, the vertical <italic>δ</italic><sup>13</sup>C gradient in all ocean basins falls between 0.8‰–1.5‰ and is consistent with that in the modern ocean, and this gradient seems to gradually increase from the shelf to the deep ocean possibly due to higher efficiency of the biological carbon pump in the open ocean commonly characterized by oligotrophy; (2) the global ocean vertical <italic>δ</italic><sup>13</sup>C gradient collapses after the K-Pg event, suggesting drastic changes in primary and/or export production in the euphotic zone; (3) the vertical <italic>δ</italic><sup>13</sup>C gradient collapse results from the abrupt negative shift of the surface ocean <italic>δ</italic><sup>13</sup>C, and the magnitude of this negative shift gradually decreases from the continental shelf to the deep sea, which may be related to the selective extinction of shallow-water grazers, swimming organisms and plankton, as well as the limited buffering capacity of shelf environments. After the K-Pg event, the main primary producers change from larger coccolithophores to smaller picoplankton, and the more recalcitrant dissolved organic carbon is able to sink to the seafloor due to the operation of the microbial carbon pump in the photic zone, leading to sustained efficiency of the biological carbon pump. Although the picoplankton-dominated biological carbon pump is not as efficient as that by coccolithophores, it is still capable of removing enough nutrients and harmful substances from the photic zone via long-term operation, not only providing food for higher trophic level organisms, but improving the marine physicochemical environment that deteriorated after the K-Pg asteroid impact. This process paves the way for the restoration and development of the entire marine ecosystem. The switch of key marine primary producers and the controlling factors involved are critical scientific issues that deserve more focus in future studies of the deep-time marine carbon cycle. The regional or global variation patterns in the biological carbon pump after the K-Pg event reveal how different marine ecosystems operate under extreme environmental stress, and provide a reference for better understanding and predicting how they respond to the ongoing rapid global climate change.