Fullerene carbon cages can encapsulate a wide variety of atoms, ions, clusters, or small molecules inside, resulting in stable compounds with unusual structures and electronic properties. These compounds are collectively defined as endohedral fullerenes. The most studied endohedral fullerenes are those containing metal atoms or ions inside, and these are referred to as endohedral metallofullerenes (EMFs). For EMFs, the inner isolated space of the fullerene cages can lead to the stabilization of unique clusters, which are otherwise not synthetically accessible. This offers an excellent environment and opportunity for investigating the nature of previously unobserved metal-metal, metal-non-metal, and metal-fullerene interactions, which are of fundamental interest and importance. Up until now, most of the work in this field has been mainly focused on the rare-earth metals and related elements (groups II, III, and IV). The encapsulation of other elements of the periodic table could potentially lead to totally new structures and bonding motifs and to material properties beyond those of the existing EMFs. Actinides were originally explored as encapsulated elements in fullerenes when Smalley et al. ( Science 1992 , 257 , 1661 ) reported mass spectral evidence of actinide endohedral fullerenes back in 1992. However, the full characterization of these actinide endohedral fullerenes, including single crystal X-ray diffractometric analyses, was not reported until very recently, in 2017. In this Account, we highlight some recent advances made in the field of EMF compounds, focusing primarily on the molecular and electronic structures of novel actinide-based EMFs, new evidence for the formation mechanisms of EMFs, and the influence of the entrapped species on the reactivity and regiochemistry of EMF compounds. We recently reported that some monometallic actinide EMFs represent the first examples of tetravalent metals encapsulated inside fullerenes that exhibit considerably stronger host-guest interactions when compared to those observed for the lanthanide EMFs. These unusually strong metal-cage interactions, along with very high mobilities of the actinides inside the fullerene cages at high temperatures, result in the stabilization of unexpected non-IPR (isolated pentagon rule) fullerene cages encapsulating only one metal ion. Strikingly, such covalent stabilization factors had never been previously observed, although Sm@C2v(19138)-C76 was the first reported mono-EMF with a non-IPR cage, see details below. In addition, we showed that a long sought-after actinide-actinide bond was obtained upon encapsulation of U2 inside an Ih(7)-C80 fullerene cage. More interestingly, we demonstrated that actinide multiple bonds, which are very difficult to prepare by conventional synthetic methods, are stabilized when trapped inside fullerene cages. A totally unexpected and previously unreported uranium carbide cluster, U═C═U, was fully characterized inside an EMF, U2C@Ih(7)-C80, which, for the first time, clearly exhibits two unsupported axial U═C double bonds that are ∼2.03 Å long. We also discovered that synthetic bis-porphyrin nanocapsules exhibit exquisitely selective complexation of some of these uranium endohedral compounds, providing the basis for a nonchromatographic EMF purification method for actinide EMFs. Regarding EMF formation mechanisms, we suggested that novel carbide EMF structures, that is, Sc2C2@Cs(hept)-C88, are likely key intermediates in a bottom-up fullerene growth process. Additionally, the structural correlation between chiral carbon cages during a bottom-up growth process was shown to be enantiomer-dependent. The influence of the encapsulated clusters on the chemical reactivity of EMFs is discussed at the end, which showed that the regioselectivities of multiple additions to the fullerene cages are remarkably controlled by the encapsulated metal clusters.