The Neogene (~23.0 – 2.6 million years ago) is the current climate analogue used in predictive scenario modeling, as it is characterized as having similar biota, geography, and environments when compared to present-day. Parameters that cannot be directly measured from the past can be estimated by stable isotope analysis measurements from proxies. Unlike most marine proxies, shark teeth provide ‘snapshots’ of environmental conditions during formation. Shark teeth can be used to calculate sea temperatures the teeth formed in after measuring δ18O of phosphate (PO4) and carbonate (CO3) within the enameloid and using δ18O of seawater (δ18Osw) estimates. Here, we model δ18OPO4 and δ18OCO3 from published linear regression equations to assess the fidelity of local T and δ18Osw being recorded in modern shark teeth. We also developed a Bayesian regression model to estimate the probability of T and δ18Osw for basins using δ18OPO4 of fossilized Neogene shark teeth from our collection and published datasets. Modern empirical δ18O and predictive δ18O* values indicate that carbonate (mean = 27.4 ± 1.5‰) is not a reliable recorder and therefore should not be considered as a paleothermometer until further constrained, but phosphate (mean = 23.5 ± 0.7‰) δ18OPO4 values were similar between taxa at localities and suggest a latitudinal temperature gradient. Variation within and between taxa may be due to species specific migration and mesothermy. Neogene T and δ18Osw estimates reflect a warmer climate, and salinity and temperature differences between the Miocene and Pliocene epochs.