Artificial intelligence (AI) has brought about profound changes in the functioning of various fields such as healthcare, industry, commerce, and finance, as well as in consumer and client behavior. To make the most of technological advancements, companies that care about customer satisfaction have started to implement product recommendation systems (RS) based on (AI). This article aims to provide a systematic literature review that can offer researchers and companies new AI methods used in product recommendation systems. Our first objective focuses on listing and understanding the different recommendation algorithms. And the second objective focuses on the evaluation techniques for the quality of (RS). The results of this study show the existence of three classic recommendation algorithms, namely collaborative filtering (CF), content-based methods, and hybrid methods. These traditional algorithms still suffer from many issues, that’s why it is important to use new (AI) methods such as recurrent neural networks, R-Transformer, Blockchain, and robo-advisors. Regarding the evaluation of the quality of the results obtained from the chosen method for the implementation of the (RS), most researchers use three indicators to assess the accuracy and completeness between the products and the clients, namely precision, recall, and the F1 score.