We compare coincident thermospheric neutral wind observations made by the Michelson Interferometer for Global High-Resolution Thermospheric Imaging (MIGHTI) on the Ionospheric Connection Explorer (ICON) spacecraft, and four ground-based specular meteor radars (SMRs). Using the green-line MIGHTI channel, we analyze 1158 coincidences between Dec 2019 and May 2020 in the altitude range from 94 to 104 km where the observations overlap. We find that the two datasets are strongly correlated (r = 0.82) with a small mean difference (4.5 m/s). Although this agreement is good, an analysis of known error sources (e.g., shot noise, calibration errors, and analysis assumptions) can only account for about a quarter of the disagreement variance. The unexplained variance is 27.8% of the total signal variance and could be caused by unknown errors. However, based on an analysis of the spatial and temporal averaging of the two measurement modalities, we suggest that some of the disagreement is likely caused by temporal variability of the wind on scales ≲70 min. The observed magnitudes agree well during the night, but during the day, MIGHTI observes 16%-25% faster winds than the SMRs. This remains unresolved but is similar in certain ways to previous SMR-satellite comparisons. Plain Language Summary Although Earth's atmosphere becomes less dense at high altitudes where it transitions to space, the wind speed grows faster, often exceeding 100 m/s (225 mph). One barrier to better predictions of conditions in the near-Earth space environment is obtaining knowledge of the wind in the thermosphere, the uppermost layer of the atmosphere. Measurements of the thermospheric wind are difficult to make and historically sparse. ICON, a new NASA mission launched in October 2019, carries the MIGHTI instrument to measure the wind from 90 to 300 km altitude. In this study we compare the observations of MIGHTI to those of meteor radars, which measure the wind from the ground by analysis of radio waves reflected by meteor trails. The results indicate good agreement between the datasets when they measure the wind at the same time and place. Specifically, with 1158 coincidences over the first 6 months of the ICON mission, the correlation is 0.82 and the average difference is 4.5 m/s. This study is important because it validates the MIGHTI data, giving confidence for subsequent studies using its data. It also quantifies limits to the agreement between space-based and ground-based winds, which is useful information for future studies combining them. HARDING ET AL.