Abstract Body (Do not enter title and authors here): Efforts to ameliorate rural healthcare burnout stemming from overwork have taken many approaches, including the increased usage of Artificial Intelligence (AI) to aid in generating electronic medical records (EMR). This new application of AI technology demands characterization of the possible benefits and efficacy of its new use. Using a commercially available AI software (Freed AI) designed for EMR production, we conducted a study to determine the quality of EMR generated by AI.

Methods
This trial analyzed 248 patient-provider interactions within a rural setting of Medicare’s Chronic Care Management (CCM program), recorded by Freed AI and a human scribe. Three blinded readers with a clinical background were given 2 notes from the same patient-encounter. One note was written by a provider and one was generated by Freed AI. Each reader performed a binary, independent assessment of each note compared with the note generated by the human control. Readers were trained on definitions of the categories clarity, accuracy, completeness, and relevance. The time required per encounter was also recorded and analyzed.

Discussion
As seen in Table 1, AI performed better than a human-generated EMR in the fields of clarity and relevance, suggesting effective formatting and a well calibrated understanding of what information is medically relevant. No significant difference in completeness was observed, suggesting AI was able to record as much valuable information as traditional charting. AI performed below the human standard in the category of accuracy. For example, AI sometimes misspelled names or misunderstood complex situations being discussed because it was not able to infer certain information which a human, using outside knowledge and past experiences, could. The reduction in time spent while utalizing AI demonstrates a highly significant decrease in workload and allows providers freedom to allocate more time interacting with each patient and hopefully reducing physician burnout.