Potato Virus Y (Potyviridae, PVY) has resulted in significant economic harm to potato (Solanum tuberosum) farmers and has disrupted seed supplies to commercial growers, especially in varieties with beneficial processing and marketing attributes but high disease susceptibility such as Russet Burbank and Russet Norkotah varieties. Commercial growers rely entirely on seed producers and certification systems to get disease-free seed as they have no recourse to mitigate seed-borne PVY after the seed is planted. Potato seed stock producers currently utilize intensive pesticide applications to suppress insect vectors and human-resource intensive activities where workers visually inspect and remove suspect infected plants during the growing season. Industry stakeholders also depend upon extensive field and tuber sampling coupled with off-season growouts and laboratory testing to ascertain infection levels within certification programs. Despite these efforts, seed producers and certification agencies are currently unable to control PVY infection in the industry’s seed pipeline, and this has a significant impact on commercial markets and regional economies. The industry also lacks a consistent, scalable, accurate, and robust detection system capable of assessing every plant within the seed potato production agro-ecosystem during the production season. Remote sensing technologies coupled with machine learning classifiers are a significant leap forward in the detection and differentiation of plant disease incidence. The continuing advancement of unmanned aerial systems (UAS) and unmanned ground vehicles provide access to high spatial, temporal, and spectral resolution instrumentation with which to monitor dynamic agricultural production systems at a leaf-scale resolution. In this study, we demonstrate PVY-infected potato plants in an agricultural production field produce different spectral reflectance profiles in comparison to neighboring non-infected plants. The Support Vector Machines (SVM) classifier differentiated spectral reflectance curves of PVY-infected and non-infected plants at an accuracy of 89.8% using near infrared and shortwave infrared wavelengths. The classification accuracy dropped to 46.9% using red, green, and blue wavelengths that the industry currently utilizes to detect diseased plants visually. This research shows that remote sensing and machine learning classifiers outperform current industry standards and have the potential to greatly enhance PVY detection efforts resulting in improved potato seed stock quality necessary to maximize yields.