The recently discovered K-Fe-Se high-temperature superconductor has caused heated debate regarding the nature of its parent compound. Transport, angle-resolved photoemission spectroscopy, and STM measurements have suggested that its parent compound could be insulating, semiconducting, or even metallic [M. H. Fang, H.-D. Wang, C.-H. Dong, Z.-J. Li, C.-M. Feng, J. Chen, and H. Q. Yuan, Europhys. Lett. 94, 27009 (2011); F. Chen et al., Phys. Rev. X 1, 021020 (2011); and W. Li et al., Phys. Rev. Lett. 109, 057003 (2012)]. Because the magnetic ground states associated with these different phases have not yet been identified and the relationship between magnetism and superconductivity is not fully understood, the real parent compound of this system remains elusive. Here, we report neutron-diffraction experiments that reveal a semiconducting antiferromagnetic (AFM) phase with rhombus iron vacancy order. The magnetic order of the semiconducting phase is the same as the stripe AFM order of the iron pnictide parent compounds. Moreover, while the sqrt[5]×sqrt[5] block AFM phase coexists with superconductivity, the stripe AFM order is suppressed by it. This leads us to conjecture that the new semiconducting magnetic ordered phase is the true parent phase of this superconductor.