I agree with the sentiment. I work on computer vision systems that have application to this area and there's a line beyond which I won't go. I'm not sure wher

I agree with the sentiment. I work on computer vision systems that have application to this area and there's a line beyond which I won't go. I'm not sure where it is exactly. For instance, there are researchers working on human detection for these systems, which can have applications in attacking and in avoiding; the tech is independent of intent. IMHO the discipline of robotics is facing some of the same agonizing choices that atomic physics faced in the 1930s-1960s. The moral dimension of their work took the physicists largely by surprise. It would be a naive roboticist who did not see the same questions coming this time.