It is imperative to identify and classify carcinogens to aid in the rapid progression of cancer epidemiology. In response to this need, O’Neill et al. have developed CarD-T, or Carcinogen Detection via Transformers, framework. This framework utilizes both transformer-based machine learning and probabilistic analysis, analyzing scientific texts and nominating potential carcinogens from the texts. Having been trained on 60% of known carcinogens, the CarD-T system was able to correctly identify the remaining known carcinogens and has additionally nominated approximately 1,600 potential carcinogens. When comparatively assessed to GPT-4, CarD-T demonstrates slightly lower precision (0.896 V.S. 0.903) but significantly stronger recall (0.853 V.S. 0.757). CarD-T is locally deployable, computationally inexpensive, comparable in precision, and stronger in recall when compared to other systems, making it an effective tool for identifying potential carcinogens in biomedical literature. With the rapidly increasing biomedical literature implicating various substances as carcinogens, a framework like this can aid scientists in upkeep with and identification of potential offenders.