Persistence of malignant clones following cytotoxic or targeted therapy is a major determinant of adverse outcome in patients with hematologic malignancies. Despite the fact that the majority of patients with acute myeloid leukemia (AML) achieve complete remission after chemotherapy, a large proportion of them relapse as a result of residual malignant cells. After discontinuation of treatment, these persistent clones have a competitive advantage and re-establish disease. Therefore, targeting strategies that specifically reduce competitive advantage of malignant cells while leaving normal cells unaffected are clearly warranted. Recently, our group identified cold-shock protein and splicing factor YBX1 as a mediator of disease persistence in JAK2-mutated myeloproliferative neoplasia. The role of cold-shock proteins in AML, however, remained so far elusive. Using genetic screening, we identified YBX1 as a relevant functional dependency in AML. Inactivation of YBX1 in vitro and in vivo confirmed its role as an essential driver of leukemia development and maintenance. Here, we identified its ability to bind specific mRNAs, including MYC, and amplify their translation at the ribosomes by recruitment to polysomal chains. Genetic inactivation of YBX1 disrupted this regulatory circuit and displaced a number of oncogenic drivers from polysomes, with subsequent depletion of protein abundance. As a consequence, leukemia cells showed reduced proliferation and were out-competed in vitro and in vivo, while normal cells remained largely unaffected. Collectively, this data establishes YBX1 as a specific dependency and therapeutic target in AML that is essentially required for oncogenic protein expression.