The possible use of <sup>14</sup>CO measurements to constrain hydroxyl radical (OH) concentrations in the atmosphere is investigated.<sup>14</sup>CO is mainly produced in the upper atmosphere from cosmic radiation. During transport to measurement locations at the Earth's surface <sup>14</sup>CO is oxidized by OH. In this paper, the sensitivity of <sup>14</sup>CO mixing ratio measurements to the 3-D OH distribution is assessed with the TM5 model. Simulated <sup>14</sup>CO mixing ratios compare reasonably well with atmospheric measurements taken at five locations worldwide. As a next step, the sensitivity of <sup>14</sup>CO measurements to OH is calculated with the adjoint TM5 model. For our sensitivity calculations the adjoint methodology outlined in the paper offers computational advantages compared to forward model calculations. The results indicate that <sup>14</sup>CO measurements, especially those taken in the tropics, are sensitive to OH in a spatially confined region. Moreover, the OH sensitivity at a certain location varies strongly over time due to meteorological variability. On average, <sup>14</sup>CO measurements are about 5 times more sensitive to OH at high latitudes than to OH in the tropics. Moreover, the measurements are sensitive to OH in the main <sup>14</sup>CO source region in the upper atmosphere. It will therefore be difficult to assign model-measurement discrepancies to either the <sup>14</sup>CO source uncertainty or to the OH sink. Nevertheless, the large distance between the <sup>14</sup>CO source region and the measurement locations should allow the retrieval of information on OH. Specifically, the sensitivity to OH in the lower atmosphere during a relatively short time span may offer the possibility to constrain local OH. These efforts will strongly depend on the number of measurements available and on our ability to accurately model the <sup>14</sup>CO transport.