Global static gravity field models, expressed in terms of spherical harmonic coefficients, are accompanied by uncertainty estimates that are critical for a wide range of geodetic and geophysical applications. Reliable error characterization is particularly important for applications such as orthometric height determination using GNSS/levelling, where geoid models derived from static gravity field models for long-wavelength components must be supported by accurate uncertainty information. Furthermore, high-precision gravity field modeling depends on appropriate weighting of satellite and terrestrial data, which in turn requires realistic, frequency-dependent noise models. Despite significant advancements in global gravity field modeling, independent validation of the provided error estimates remains limited. In this study, we present a novel framework for validating uncertainty estimates of static gravity field models using Line-of-Sight Gravity Difference (LGD) observations derived from the GRACE-FO Laser Ranging Interferometer (LRI). Residual LGD LRI time series referenced to a specific static gravity field model are obtained by removing static as well as time-variable gravity signals using monthly Level-2 gravity field solutions developed at the University of Colorado Boulder. The spectral and spatial characteristics of the resulting LRI residuals referenced to different static gravity field models are analyzed and compared with model-propagated error estimates. The proposed validation framework is applied to assess the reliability of the error provided by several widely used global static gravity field models, including both satellite-only and combined solutions. The results reveal discrepancies in specific frequency bands as well as spatial regions, indicating under- or over-estimated uncertainties of the models. In particular, the findings emphasize the critical role of full covariance information in accurately capturing the spatial structure of the model errors. Overall, this work establishes LGD observations from GRACE-FO LRI as a powerful tool for the independent validation of uncertainty estimates in global static gravity field models.