Diabetic retinopathy (DR) is a complication of diabetes that can lead to blindness and visual impairment in adults. Timely clinical intervention, particularly through laser treatment, can reduce significant vision loss by up to 90%, effectively preventing further deterioration. Current recommendations advocate for annual ophthalmology check-ups specifically focused on DR for all individuals with diabetes. This examination involves capturing images using a fundus camera and subsequently interpreting them. The initial step is performed by an orthoptist, while the second is delegated to an ophthalmologist. This latter stage presents a considerable workload for ophthalmologists, a demand that is projected to rise in the coming years due to the relentless increase in the diabetic patient population. While Artificial Intelligence (AI) finds broad applications across various medical domains, ophthalmology stands out as a field that extensively embraces AI technology. It continues to be at the forefront of utilizing AI for the detection, diagnosis, and treatment of various ocular conditions, including diabetic retinopathy. This advancement holds the potential to enhance patient access to care and may contribute to addressing challenges in the healthcare system. The automated detection of pathologies associated with DR primarily relies on the extraction of key anatomical structures within the retina, including the vascular network, optic papilla, and macula. The objective of this review is to provide an overview of the most recent AI-based DR screening methods discussed in the literature, some of which have already entered the market. These technologies have been developed utilizing diverse technical approaches and training datasets.