The duplicated and the highly repetitive nature of the maize genome has historically impeded the development of true single nucleotide polymorphism (SNP) markers in this crop. Recent advances in genome complexity reduction methods coupled with sequencing-by-synthesis technologies permit the implementation of efficient genome-wide SNP discovery in maize. In this study, we have applied Complexity Reduction of Polymorphic Sequences technology (Keygene N.V., Wageningen, The Netherlands) for the identification of informative SNPs between two genetically distinct maize inbred lines of North and South American origins. This approach resulted in the discovery of 1,123 putative SNPs representing low and single copy loci. In silico and experimental (Illumina GoldenGate (GG) assay) validation of putative SNPs resulted in mapping of 604 markers, out of which 188 SNPs represented 43 haplotype blocks distributed across all ten chromosomes. We have determined and clearly stated a specific combination of stringent criteria (>0.3 minor allele frequency, >0.8 GenTrainScore and >0.5 Chi_test100 score) necessary for the identification of highly polymorphic and genetically stable SNP markers. Due to these criteria, we identified a subset of 120 high-quality SNP markers to leverage in GG assay-based marker-assisted selection projects. A total of 32 high-quality SNPs represented 21 haplotypes out of 43 identified in this study. The information on the selection criteria of highly polymorphic SNPs in a complex genome such as maize and the public availability of these SNP assays will be of great value for the maize molecular genetics and breeding community.