Marine dimethyl sulfide (DMS), a biogenic element in the ocean, is important to the climate due to its ability to alter Earth’s radiation budget. Compared to its concentration in the atmosphere, DMS in the surface ocean is supersaturated. Being released from the ocean to the atmosphere through the air-sea gas exchange, DMS imposes a direct effect on the atmospheric environment. Furthermore, the sulfate aerosol formed by the oxidation of DMS is the primary source of cloud condensation nuclei, which is conducive to the formation of clouds. Therefore, it helps increase the reflection efficiency of the earth (albedo), thus leading to a negative greenhouse effect. In this study, we train an artificial neural network model using observations of DMS and two environmental parameters, sea surface ocean temperature (SST) and sea surface ocean salinity (SSS). To make a future prediction, we apply the model to parameters extracted from the simulations of the Coupled Model Intercomparison Project 6 (CMIP6). Our results show that DMS concentration and sea-to-air flux overall decrease with a warming climate, indicating positive feedback.