Lysine deacetylases (KDACs or HDACs) are metal-dependent enzymes that regulate lysine acetylation, a post-translational modification that is present on thousands of human proteins, essential for many cellular processes, and often misregulated in diseases. The selective inhibition of KDACs would allow for understanding of the biological roles of individual KDACs and therapeutic targeting of individual enzymes. Recent studies have suggested that purportedly specific KDAC inhibitors have significant off-target binding, but the biological consequences of off-target binding were not evaluated. We compared the effects of treatment with two of the reportedly most KDAC-selective inhibitors, Tubastatin A and PCI-34051, in HT1080 cells in which the endogenous KDAC6 or KDAC8 gene has been mutated to inactivate enzyme catalysis while retaining enzyme expression. Genetic inactivation results in much stronger deacetylation defects on known targets compared to inhibitor treatment. Gene expression analysis revealed that both inhibitors have extensive and extensively overlapping off-target effects in cells, even at low inhibitor doses. Furthermore, Tubastatin A treatment led to increased histone acetylation, while inactivation of KDAC6 or KDAC8 did not. Genetic inactivation of KDAC6, but not KDAC8, impaired tumor formation in a xenograft model system, in contrast to previous reports with KDAC inhibitors suggesting the reverse. We conclude that the majority of observed biological effects of treatment with KDAC inhibitors are due to off-target effects rather than the intended KDAC inhibition. Developing a truly specific KDAC6 inhibitor could be a promising therapeutic avenue, but it is imperative to develop new inhibitors that selectively mimic genetic inactivation of individual KDACs.