You say: to increase acceptance, to accomplish social tasks, and to motivate humans interactions.
I add, to help humans like social assitive robots. In this context, social robots have been generating positives effects in humans. For example, in healthcare and education contexts.
Hi Patrick! They are really good points. I have been discussing the benefits of social interfaces in service robots for some years with colleagues and I think the main barriers are the design approaches used when the robots are conceived. Adding to your points, I think they could help to mediate Human-Human interaction in different setups. I hope this answer is useful for you.
I would like to add to this question how Social robotics compare in this terms with current technologies in order to frame in a specific way these point you describe in your question.
Is the question about benefits in a narrow sense of economic gain, or in the larger sense of serving human well-being? If it is meant in the latter sense, then the "non-replacement principle "of ISR (Integrative Social Robotics) can offer a heuristics to identfy beneficial applications: "social robots may only do what humans should but cannot do".