The field of digital health technology for diabetes management is moving towards increased use of artificial intelligence (AI) and machine learning (ML) to improve patient outcomes. A key challenge in this area is the lack of access to high-quality datasets, which is being addressed through the creation of publicly available collections of longitudinal diabetes data. These datasets are being used to develop and evaluate AI algorithms for tasks such as blood glucose prediction, and are providing valuable insights into the performance of different algorithms and the importance of dataset selection. Another area of focus is the automation of data extraction and analysis, with recent studies evaluating the performance of large language models for tasks such as meta-analysis data extraction. While these models show promise, they also have limitations, and guidelines are being proposed for their use in real-world applications. Noteworthy papers include Glucose-ML, which presents a collection of 10 publicly available diabetes datasets, and a study on the use of large language models for meta-analysis data extraction, which proposes a three-tiered set of guidelines for using these models.