Recent studies have uncovered the potential of Mg isotopes (δ26Mg) for studying past ocean chemistry, but records of such data are still scarce. Dolomite has been suggested as a promising archive for δ26Mg of seawater. However, its enigmatic formation mechanism and the difficulty in precipitating dolomite in the laboratory at surface temperatures decrease confidence in the interpretation of δ26Mg values from the rock record. To evaluate factors determining the δ26Mg of dolomite, we studied pore water and sediment from Dohat Faishakh Sabkha, Qatar—one of the rare environments where dolomite is currently forming. The δ26Mg values of the dolomite (−2.56‰ to −1.46‰) are lower than that of seawater (−0.83‰), whereas δ26Mg values of pore water (−0.71‰ to −0.14‰) are higher. The isotope fractionation accompanying dolomite formation is generally in accordance with an empirical fractionation from the literature, extrapolated to the sabkha’s temperature (−1.84‰ to −1.51‰). The results suggest that evaporated seawater is the sole source of Mg, and isotopically light dolomite is the major sink, so that the δ26Mg of the dolomite-forming pore water is equal to or greater than that of seawater. Thus, provided that the lowest δ26Mg value among several dolomite samples is used, and the formation temperature is known, similar sabkha-type dolomites can be utilized as an archive for δ26Mg values of ancient seawater.