This study used the Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) 22-item checklist to evaluate predictive AI models in Spine Surgery. Articles were identified through PubMed searches for Spine surgery-specific predictive AI models in the top six orthopedic and neurosurgery journals. Using the TRIPOD checklist, reviews were performed to assess studies based on their ability to meet TRIPOD criteria. 72 articles were reviewed in total, with the median compliance to the TRIPOD criteria being 57.14%. The highest compliance achieved was 93.10%, with pediatric spine surgery achieving the highest median compliance at 66.66%. The data provided in the study elaborates on the exact details of AI model types, publication year, study type, and journals published in. Overall, the study found an inadequate compliance according to the TRIPOD criteria, suggesting that further studies should be more carefully developed before healthcare providers can implement any AI models. In particular, the median compliance in the methods and results sections was 33.33%, meaning limited generalizability. The study also acknowledges its limitations in reviewing papers through the top six journals, which could lower the generalizability.