To assess the tools for artificial intelligence (AI) in evaluating reproductive medicine. This is a study to explore the latest trends of AI for reproductive medicine.

From the on line website (https://www.ncbi.nlm.nih.gov) available, we used keywords of reproductive medicine and AI published articles in recent years (between 2023 and March, 2025) for survey and for further analysis.

There were one hundred and sixty-five articles found at the year of 2023. After manual filtering, there were thirty-three articles for human–based studies available for analysis. More than 88% of the articles were dealing with oocyte-related conditions at the year of 2023. Within one year up to Mrach,2025, there were thirty-seven human-based articles filtered from seventy-six published articles. Only four articles (10%) dealt with male side (semen or sperm-related). It showed that two categories of AI methods could be used: machine learning (ML), and natural language processing (NLP). There were three ways of ML method: supervised learning, unsupervised learning and reinforcement learning. Supervised learning was more widely used in reproductive medicine. In 2025, the implementation of generative AI marked the initial phase of enhancing the robustness of AI algorithms. AI could be feasible and helpful to reproductive medicine in several ways: oocytic evaluation and selection; selection of semen or sperm quality; embryo selection which may raise ethical issues right away; and prediction of the outcome of in vitro fertilization (IVF). As of the submission date, commercially available or quality-assured software algorithms for daily clinical practice remained unavailable.

In the current survey, not all AI methods are practical or readily applicable. AI remains in its early developmental stages. However, development may by speedy in the near future. Further long term survey is needed in AI of reproductive medicine.