Small, negatively charged gold clusters isolated in vacuum can oxidize CO via electron-transfer-mediated activation of O2. This suggests that Au clusters can act as aerobic oxidation catalysts in the real world when their structure parameters satisfy given required conditions. However, there is a technical challenge for the development of Au cluster oxidation catalysts; the structural parameters of the Au clusters, such as size and composition, must be precisely controlled because the intrinsic chemical properties of the clusters are strongly dependent on these parameters. This Account describes our efforts to achieve precision synthesis of small (diameter <2 nm) Au clusters, stabilized by polymers and immobilized on supports, for a variety of catalytic applications. Since we aim to develop Au cluster catalysts by taking full advantage of their intrinsic, size-specific chemical nature, we chose chemically inert materials for the stabilizers and supports. We began by preparing small Au clusters weakly stabilized by polyvinylpyrrolidone (PVP) to test the hypothesis that small Au clusters in the real world will also show size-specific oxidation catalysis. The size of Au:PVP was controlled using a microfluidic device and monitored by mass spectrometry. We found that only Au clusters smaller than a certain critical size show a variety of aerobic oxidation reactions and proposed that the reactions proceed via catalytic activation of O2 by negatively charged Au clusters. We also developed a method to precisely control the size and composition of supported Au clusters using ligand-protected Au and Au-based bimetallic clusters as precursors. These small Au clusters immobilized on mesoporous silica, hydroxyapatite, and carbon nanotubes acted as oxidation catalysts. We have demonstrated for the first time an optimal Au cluster size for the oxidation of cyclohexane and a remarkable improvement in the oxidation catalysis of Au25 clusters by single-atom Pd doping. The non-scalable catalysis of Au clusters that we reported here points to the possibility that novel catalysis beyond that expected from bulk counterparts can be developed simply by reducing the catalyst size to the sub-2 nm regime.