The field of computer vision is moving towards more accurate and robust detection of small and rare species in aerial imagery, with a focus on multi-scale consistency and context-aware augmentation. Researchers are also exploring the limitations of generative image models, including geographic knowledge and diversity deficits, as well as conceptual blindspots. Noteworthy papers include RareSpot, which proposes a robust detection framework for small and rare wildlife, and OpenWildlife, which introduces an open-vocabulary wildlife detector for multi-species identification. Additionally, Uncovering Conceptual Blindspots in Generative Image Models Using Sparse Autoencoders presents a systematic approach for identifying and characterizing conceptual blindspots in generative image models.