Forecasting home prices is very challenging due to the large number of variables that affect this dynamic market. The complexity lies in the interaction between economic, social, and political factors that together determine the value of the property. Price setting on a real estate property depends on a myriad of variables, some of which are quantitative, while others are qualitative. The former can probably be measured or estimated, while the latter are mainly guessed by an individual who has the appropriate expertise. But bearing in mind the importance of context, subjectivity and the dynamism of the market, it can be easily understood that prices of real estate properties are almost impossible to predict. Unless AI techniques, that gather vast sums of information, are in use. The present paper concerns the question of whether price formation in the real estate sector can be predicted by using AI techniques. Due to the novelty of the topic, a qualitative methodology based on a bibliometric analysis has been put in place. The results show a rapid evolution of research that has accelerated in recent years, adopting new terminology as technology develops. This is associated with the participation of different relevant authors that form a decentralized knowledge network.