Alliance-level vegetation classification methods are typically used for landscape scale assessments and accordingly may not represent the exact species composition of the ground cover on a fine scale. This methodology contains opportunity for error related to human influence, field limitations, and/or recording of spatial data. Human-related error may include inadvertent biases or subjectivity, such as differences in classification methods or measurements between surveyors. By leveraging emerging technologies such as artificial intelligence (AI), sources of human error and survey area constraints may be largely eliminated. In 2022, Understory (formerly Comon Solutions) contracted Sequoia Ecological Consulting, Inc. (Sequoia) to provide field services and data analysis in support of the Doty Ravine Pilot Study, an effort to train Understory’s AI algorithm on plant species identification as well as vegetation classification across a landscape. Sequoia collected full-coverage vegetation classification using typical, standardized field techniques; captured near-simultaneous, high-resolution aerial imagery with small, unmanned aerial systems (sUAS); compared the results of different data capture methods; and provided recommendations for integration of future sampling efforts to continue refinement of training methods and the algorithm itself. Algorithmic data showed agreement with botanist-defined alliance polygons, and the study results provided insight on improving validation methods.