Natural dissolved organic matter (DOM) represents a ubiquitous molecular mixture, progressively characterized by spatiotemporal resolution. However, an inadequate comprehension of DOM molecular dynamics, especially the stochastic processes involved, hinders carbon cycling predictions. This study employs ecological principles to introduce a neutral theory to elucidate the fundamental processes involving molecular generation, degradation, and migration. A neutral model is thus formulated to assess the probability distribution of DOM molecules, whose frequencies and abundances follow a β-distribution relationship. The neutral model is subsequently validated with high-resolution mass spectrometry (HRMS) data from various waterbodies, including lakes, rivers, and seas. The model fitting highlights the prevalence of molecular neutral distribution and quantifies the stochasticity within DOM molecular dynamics. Furthermore, the model identifies deviations of HRMS observations from neutral expectations in photochemical and microbial experiments, revealing nonrandom molecular transformations. The ecological null model further validates the neutral modeling results, demonstrating that photodegradation reduces molecular stochastic dynamics at the surface of an acidic pit lake, while random distribution intensifies at the river surface compared with the porewater. Taken together, the DOM molecular neutral model emphasizes the significance of stochastic processes in shaping a natural DOM pool, offering a potential theoretical framework for DOM molecular dynamics in aquatic and other ecosystems.