Falciparum malaria is initiated when Anopheles mosquitoes transmit the Plasmodium sporozoite stage during a blood meal. Irradiated sporozoites confer sterile protection against subsequent malaria infection in animal models and humans. This level of protection is unmatched by current recombinant malaria vaccines. However, the live-attenuated vaccine approach faces formidable obstacles, including development of accurate, reproducible attenuation techniques. We tested whether Plasmodium falciparum could be attenuated at the early liver stage by genetic engineering. The P. falciparum genetically attenuated parasites (GAPs) harbor individual deletions or simultaneous deletions of the sporozoite-expressed genes P52 and P36. Gene deletions were done by double-cross-over recombination to avoid genetic reversion of the knockout parasites. The gene deletions did not affect parasite replication throughout the erythrocytic cycle, gametocyte production, mosquito infections, and sporozoite production rates. However, the deletions caused parasite developmental arrest during hepatocyte infection. The double-gene deletion line exhibited a more severe intrahepatocytic growth defect compared with the single-gene deletion lines, and it did not persist. This defect was assessed in an in vitro liver-stage growth assay and in a chimeric mouse model harboring human hepatocytes. The strong phenotype of the double knockout GAP justifies its human testing as a whole-organism vaccine candidate using the established sporozoite challenge model. GAPs might provide a safe and reproducible platform to develop an efficacious whole-cell malaria vaccine that prevents infection at the preerythrocytic stage.