Genomic selection (GS) has transformed plant breeding by enabling more accurate prediction of breeding values using genome-wide marker data. Traditional marker-assisted selection (MAS) has been ineffective for complex traits influenced by many genes with small effects, but GS uses all marker data to predict performance, leading to more accurate and efficient selection. GS relies on a training population of genotyped and phenotyped individuals to develop a model that predicts genomic estimated breeding values (GEBVs) for untested individuals. These GEBVs are ideal for selection, as they do not depend on the function of underlying genes but rather on the genetic relationships between individuals. GS has shown remarkable accuracy in simulations and empirical studies across various species, including dairy cattle, mice, and plants. It outperforms traditional selection methods, especially when marker density is high and when genetic relationships are strong. However, the effectiveness of GS depends on factors such as marker density, training population size, and the genetic architecture of the trait. Methods like ridge regression and BayesB have been proposed to improve prediction accuracy, with BayesB often performing better when markers are strongly associated with QTL. Theoretical studies highlight that GS accuracy is influenced by both linkage disequilibrium (LD) between markers and QTL and genetic relationships between individuals. GS can maintain genetic diversity while increasing selection gains, as it allows for more accurate predictions of breeding values. However, the long-term impact of GS on genetic diversity and selection response depends on the balance between these factors. Empirical studies in livestock, such as dairy cattle, have demonstrated the effectiveness of GS in improving breeding values. These studies show that GS can outperform traditional selection methods, especially when marker data is available. The application of GS in plant breeding is also promising, with studies showing that it can enhance the efficiency and accuracy of selection for complex traits. Future research needs to focus on optimizing training population design, understanding the genetic architecture of traits, and improving prediction methods. GS has the potential to revolutionize plant breeding by enabling faster and more accurate selection, but further theoretical and empirical studies are needed to fully realize its potential.Genomic selection (GS) has transformed plant breeding by enabling more accurate prediction of breeding values using genome-wide marker data. Traditional marker-assisted selection (MAS) has been ineffective for complex traits influenced by many genes with small effects, but GS uses all marker data to predict performance, leading to more accurate and efficient selection. GS relies on a training population of genotyped and phenotyped individuals to develop a model that predicts genomic estimated breeding values (GEBVs) for untested individuals. These GEBVs are ideal for selection, as they do not depend on the function of underlying genes but rather on the genetic relationships between individuals. GS has shown remarkable accuracy in simulations and empirical studies across various species, including dairy cattle, mice, and plants. It outperforms traditional selection methods, especially when marker density is high and when genetic relationships are strong. However, the effectiveness of GS depends on factors such as marker density, training population size, and the genetic architecture of the trait. Methods like ridge regression and BayesB have been proposed to improve prediction accuracy, with BayesB often performing better when markers are strongly associated with QTL. Theoretical studies highlight that GS accuracy is influenced by both linkage disequilibrium (LD) between markers and QTL and genetic relationships between individuals. GS can maintain genetic diversity while increasing selection gains, as it allows for more accurate predictions of breeding values. However, the long-term impact of GS on genetic diversity and selection response depends on the balance between these factors. Empirical studies in livestock, such as dairy cattle, have demonstrated the effectiveness of GS in improving breeding values. These studies show that GS can outperform traditional selection methods, especially when marker data is available. The application of GS in plant breeding is also promising, with studies showing that it can enhance the efficiency and accuracy of selection for complex traits. Future research needs to focus on optimizing training population design, understanding the genetic architecture of traits, and improving prediction methods. GS has the potential to revolutionize plant breeding by enabling faster and more accurate selection, but further theoretical and empirical studies are needed to fully realize its potential.