Antiferromagnets (AF) have long remained an intriguing and exotic state of matter, their application being restricted to enabling interfacial exchange bias in spin-valves. Their role in the expanding field of applied spintronics has been mostly passive and the in-depth investigation of their basic properties considered as fundamental condensed matter physics. A conceptual breakthrough was achieved ten years ago with the proposal that spin transfer torque could be used to electrically control the direction of the order parameter of AF spin valves, henceforth making these materials potential candidates for low energy spin devices. In spite of substantial theoretical efforts and experimental attempts to observe such a torque, the difficulty to independently detect the direction of the AF order parameter has remained a major obstacle.
In this talk, I will first introduce the original concept of spin transfer torque in AF spin-valves, demonstrating that it is strongly limited by the spin decoherence and dramatically vanishes in the presence of disorder, leaving little hope to observe this effect experimentally. Then, I will present the newly proposed concept of spin-orbit torque that utilizes bulk or interfacial the spin-orbit coupling in non-centrosymmetric magnets to directly generate a torque on the AF order parameter. This torque, being local, is much more robust against impurities, as will be demonstrated for the specific case of interfacial Rashba spin-orbit coupling. Finally, I will discuss about spin motive force and torques in antiferromagnetic textures, intriguing effects that remained to be experimentally observed.