In this project we investigate the atomic vacancies of a range of simple oxides using cutting-edge computational structure prediction and first-principles density-functional theory (DFT). The oxides are a class of materials which have thousands of technologically relevant uses: from gate dielectrics in field-effect transistors (FETs), solid-oxide fuel cells and superconductors for low-loss power transmission. Furthermore, some oxides also piezo- or ferroelectric making them useful for microelectromechanical systems (MEMS). The ab initio random structure searching method has found new low-energy self-interstitial/hydrogen and H/N/O complexes in silicon along with charged oxygen interstitials and He ions in the complex oxide, zirconolite, (CaZrTi2O7) which is used for nuclear waste encapsulation. There is evidence that in even the simplest of oxides, such as, Na2O, MgO, Al2O3, SiO2, P2O5, new types of vacancies can still be uncovered, suggesting that novel applications could be realised from well-known materials.