Evidence suggests custom chatbots are superior to commercial generative artificial intelligence (GenAI) systems for text‐based anatomy content inquiries. This study evaluates ChatGPT‐4o's and Claude 3.5 Sonnet's capabilities to interpret unlabeled anatomical images. Secondarily, ChatGPT o1‐preview was evaluated as an AI rater to grade AI‐generated outputs using a rubric and was compared against human raters. Anatomical images (five musculoskeletal, five thoracic) representing diverse image‐based media (e.g., illustrations, photographs, MRI) were annotated with identification markers (e.g., arrows, circles) and uploaded to each GenAI system for interpretation. Forty‐five prompts (i.e., 15 first‐order, 15 second‐order, and 15 third‐order questions) with associated images were submitted to both GenAI systems across two timepoints. Responses were graded by anatomy experts for factual accuracy and superfluity (the presence of excessive wording) on a three‐point Likert scale. ChatGPT o1‐preview was tested for agreement against human anatomy experts to determine its usefulness as an AI rater. Statistical analyses included inter‐rater agreement, hierarchical linear modeling, and test–retest reliability. ChatGPT‐4o's factual accuracy score across 45 outputs was 68.0% compared to Claude 3.5 Sonnet's score of 61.5% (p = 0.319). As an AI rater, ChatGPT o1‐preview showed moderate to substantial agreement with human raters (Cohen's kappa = 0.545–0.755) for evaluating factual accuracy according to a rubric of textbook answers. Further improvements and evaluations are needed before commercial GenAI systems can be used as credible student resources in anatomy education. Similarly, ChatGPT o1‐preview demonstrates promise as an AI assistant for educational research, though further investigation is warranted.

Accurate and precise age‐at‐death estimation methods are critical when studying past lifeways. However, adult age‐at‐death estimation is often difficult because of diverse physiologies, preservation, and timing of biological processes in target and reference populations. These challenges can complicate the comparison of results between studies, which can also be impacted by the training and method preference of each bioarchaeologist. In this paper, we first compare the use of two types of age‐at‐death estimation methods, namely, traditional methods (e.g., pubic symphyseal aging) and Transition Analysis. Second, we build upon the work of Falys and Lewis (2011), who reported frequencies of adult age‐at‐death method use between 2000 and 2010, by reporting the frequency of use for traditional methods over the past 40 years with special attention on the last decade. We build further on their work by identifying trends in the minimum age at adulthood, adult age categories, geographic region of institutional affiliation of the first author, and the geographic region of sample origin and interpreting their possible implications on comparability of bioarchaeological literature. Our results show that Transition Analysis, institutional affiliation, and sample origin do not represent major sources of variation in bioarchaeological literature so far. In the case of Transition Analysis, this could result from its limited use thus far, while institutional affiliation and sample origin do not represent major sources of variation because of the relative lack of diversity in both variables. There was, however, substantial variation in the type of traditional method used, the minimum age at adulthood, and adult age categories, which could affect comparability between studies. Our study highlights the need for continued discussion about standardizing bioarchaeological age‐at‐death estimation while respecting the unique needs of diverse target samples.

Age-at-death estimation is influenced by biological and environmental factors. Physiological stress is intertwined with these factors, yet their impact on senescence and age estimation is unknown. Stature, linear enamel hypoplasia (LEH), and antemortem tooth loss (AMTL) in the Hamann–Todd Osteological Collection (n = 297) are used to understand whether physiological stress is related to age estimation inaccuracy using transition analysis (TA). Considering the low socioeconomic status of individuals in the collection, it was expected that many people experienced moderate to severe physiological stressors throughout their lives. Of the sample, 44.1% had at least one LEH, but analyses found no relationship between LEH incidence and TA error. There was no association between stature and TA error for males or females. However, females with at least one LEH had significantly shorter statures (t = 2.412, p = 0.009), but males did not exhibit the same pattern (t = 1.498, p = 0.068). Further, AMTL frequency and TA error were related (r = 0.276, p < 0.001). A partial correlation controlling for age-at-death yielded a correlation coefficient of 0.024 (p = 0.684), suggesting that this relationship is mostly explained by age-at-death. These data suggest that age estimation methods are not significantly affected by physiological stress in this sample, but further investigations are needed to understand how these variables relate to skeletal aging.