Talking to machines is not evoking science fiction scenes as it used to be in the sixties or even the nineties, any more. With the successful demonstration of industrial-strength enterprise and consumer facing dialogue systems, we now have a set of next-level questions that the community of dialogue researchers, business and consumer organizations, and policy makers are facing: What type of fundamental research in dialogue research do we need to address? How can we keep a research flow from academic labs to industry labs? What types of next-generation conversational interfaces should we develop? How can we build dialogue systems that are transparent about their capabilities? How can the user community build trust in conversational machines? What ethical rules should conversational systems follow? How do we make transparent the inherent bias in data-driven machine learning algorithms embedded in conversational systems?