Oxides have become a key ingredient for new concepts of electronic devices. To a large extent, this is due to the profusion of new physics and novel functionalities arising from ultrathin oxide films and at oxide interfaces. We present here a perspective on selected topics within this vast field and focus on two main issues. The first part of this review is dedicated to the use of ultrathin films of insulating oxides as barriers for tunnel junctions. In addition to dielectric non-magnetic epitaxial barriers, which can produce tunneling magnetoresistances in excess of a few hundred percent, we pay special attention to the possibility of exploiting the multifunctional character of some oxides in order to realize ‘active’ tunnel barriers. In these, the conductance across the barrier is not only controlled by the bias voltage and/or the electrodes magnetic state, but also depends on the barrier ferroic state. Some examples include spin-filtering effects using ferro- and ferrimagnetic oxides, and the possibility of realizing hysteretic, multi-state junctions using ferroelectric barriers. The second part of this review is devoted to novel states appearing at oxide interfaces. Often completely different from those of the corresponding bulk materials, they bring about novel functionalities to be exploited in spintronics and electronics architectures. We review the main mechanisms responsible for these new properties (such as magnetic coupling, charge transfer and proximity effects) and summarize some of the most paradigmatic phenomena. These include the formation of high-mobility two-dimensional electron gases at the interface between insulators, the emergence of superconductivity (or ferromagnetism) at the interface between non-superconducting (or non-ferromagnetic) materials, the observation of magnetoelectric effects at magnetic/ferroelectric interfaces or the effects of the interplay and competing interactions at all-oxide ferromagnetic/superconducting interfaces. Finally, we link up the two reviewed research fields and emphasize that the tunneling geometry is particularly suited to probe novel interface effects at oxide barrier/electrode interfaces. We close by giving some directions toward tunneling devices exploiting novel oxide interfacial phenomena.