Colorectal cancer (CRC) is the third most commonly diagnosed malignancy and the second leading cause of cancer related-death in the world [1]. CRC develops from precancerous polyps through several (epi)genetic pathways [2]. Colonoscopy is considered the gold standard for detection and diagnosis of CRC and its precursor lesions [3] [4]. Moreover, colonoscopy provides opportunities for endoscopic resection of precancerous polyps, which is known to be effective to prevent CRC [5]. Despite reported benefits, colonoscopy outcomes may differ depending on quality of the endoscopist performing the procedure. Among others, this relates to differences in ability of endoscopists to accurately assess polyp characteristics such as location [6], size [7] [8] [9] [10] and morphology [11] [12] [13], as well as differences in their performance in predicting histological polyp features (e.g. histological subtype [14] [15] [16], grade of dysplasia [17] [18] [19], and, in case of suspected malignancy, presence of deep submucosal invasion [DSI] [20] [21] [22]). These polyp characteristics are essential to decide on the indication for resection and histopathological analysis [23] [24], appropriate resection method [25] [26] and appropriate surveillance interval [27] [28]. Hence, inaccurate endoscopic polyp assessment could lead to higher patient and economic burden due to unnecessary polyp resection and analysis, as well as suboptimal treatment and/or surveillance regimens. Over the last decade, artificial intelligence (AI) in biomedical science has received growing attention. AI can be defined as the simulation of human intelligence by computer systems [29]. Specific AI techniques such as machine learning can be used to make machines (computers) smarter through experience-based learning [30] [31]. Since computer systems can be trained with a large amount of high quality and expert-annotated data, they could possibly serve as an objective, real-time, expert-level second observer modality during colonoscopy procedures. This might provide opportunities to reduce interobserver variability and improve general polyp assessment accuracy. While evidence is currently scattered, we aimed to write a narrative review to provide a broad overview of current developments within the field of AI and computer science for computer-aided assessment of colorectal polyps. This includes assessment of polyp location, size, morphology and histology, including degree of dysplasia (low grade dysplasia [LGD] versus high grade dysplasia [HGD]) and, in case of suspected malignancy, invasion depth. Since computer-aided polyp detection concerns an already more thoroughly studied and evaluated topic [32] [33], developments within this field will not be addressed within this review. Received: 04 February 2023 Accepted after revision: 08 May 2023 Article published online: 16 August 2023 © 2023. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/). Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany