The paper discusses the extension of pseudo-relativistic physics from two-dimensional (2D) materials like graphene to three-dimensional (3D) materials, focusing on the existence of 3D Dirac points. Unlike phase transitions from topological to normal insulators, specific space groups allow symmetry-protected 3D Dirac points. The authors provide criteria for identifying these groups and present *ab initio* calculations of β-cristobalite BiO$_2$, which exhibits three Dirac points at the Fermi level. They find that β-cristobalite BiO$_2$ is metastable and can be physically realized as a 3D analog to graphene. The paper also explores the conditions under which a 3D double space-group can host a Dirac point, including the requirement for four-dimensional irreducible representations (FDIRs) and the linear dispersion of bands around these points. The authors apply these criteria to diamond and zincblende lattices, concluding that β-cristobalite BiO$_2$ meets the conditions for hosting a Dirac semimetal. They discuss the challenges in realizing such a material in realistic systems and propose modifications to the lattice structure to avoid additional Fermi surface pockets.The paper discusses the extension of pseudo-relativistic physics from two-dimensional (2D) materials like graphene to three-dimensional (3D) materials, focusing on the existence of 3D Dirac points. Unlike phase transitions from topological to normal insulators, specific space groups allow symmetry-protected 3D Dirac points. The authors provide criteria for identifying these groups and present *ab initio* calculations of β-cristobalite BiO$_2$, which exhibits three Dirac points at the Fermi level. They find that β-cristobalite BiO$_2$ is metastable and can be physically realized as a 3D analog to graphene. The paper also explores the conditions under which a 3D double space-group can host a Dirac point, including the requirement for four-dimensional irreducible representations (FDIRs) and the linear dispersion of bands around these points. The authors apply these criteria to diamond and zincblende lattices, concluding that β-cristobalite BiO$_2$ meets the conditions for hosting a Dirac semimetal. They discuss the challenges in realizing such a material in realistic systems and propose modifications to the lattice structure to avoid additional Fermi surface pockets.