About the RCN

Members of the WAIM network are developing the understanding needed to jointly design both sides of the human-technology frontier in work settings and using intelligent machines. We use the phrase intelligent machines, focusing attention on computing technologies characterized by autonomy, the ability to learn, and the ability to interact with other systems and with humans. Intelligent machines are not capable of the generalized intelligence humans have. But they are increasingly capable of performing tasks that traditionally have been in the sole purview of humans. For example, machines can now recognize images or speech in particular domains with an ability approaching or even surpassing that of humans. These abilities enable them to perform useful work, and in some cases they can perform this work more accurately than humans, with greater speed and at less cost. Applications of these abilities are already beginning to affect labor markets and the nature of work. However, while computing technologies are becoming ever more capable, the human side of the frontier—comprising people, organizations, legal frameworks, and social values, to name a few—is evolving much more slowly. The result is an “impedance mismatch”—between the technologies and the organizational and individual abilities to incorporate automated technologies into full use—that risks unexpected or undesired consequences (e.g., deskilling, overly fragile systems or “automation surprises”). Much of the rhetoric around work and intelligent machines focuses on the possibilities of people being put out of work by automation. But this view is too simplistic a conception of how people and intelligent machines will interact, because tasks that can be automated rarely stand in isolation (Chui et al., 2015). Indeed, context often shapes tasks. As an example, it may already be feasible to develop an automated system to diagnose skin cancer (Esteva et al., 2017), but to be practicable, such a system needs to fit the complex work of a medical practice. Someone must order the imaging, image the correct area of the body using the right lighting, explain the diagnosis to the patient, family members, or other doctors (in varied and appropriate ways), bill insurance companies, monitor ongoing performance, defend malpractice suits, and so on. All this surrounding work needs to adapt to an automated dermatologist (and vice versa). As a result, analysts expect that “technological disruptions such as robotics and machine learning—rather than completely replacing existing occupations and job categories—are likely to substitute specific tasks previously carried out as part of these jobs” (World Economic Forum, 2016), leading to increased demand for the remaining skills. (Indeed, research has already noted an increase in the demand for social skills (Deming, 2015).) We argue, therefore, that system design must be addressed as a socio-technological problem, requiring the joint design of social and technological systems and attention to the implications of their interdependencies. The challenge of understanding and proactively designing work shared with (rather than simply replaced by) intelligent machines requires expanding from a delimited focus on the technical systems and their potential for autonomous action. A more expansive perspective accords with, e.g., the focus of the National Robotics Initiative 2.0 (NSF 17–518), which details the need for attention to "robots... that work beside or cooperatively with people.” Yet, most research today centers primarily on technology and its potential for autonomous action, with less attention to designing across the frontier. Furthermore, because of the need to simultaneously consider technical, individual, group, organizational, and societal issues, only convergent research can build the deep and systematic knowledge required to engage the complex questions that need to be addressed to design work that best leverages expanding technological capabilities and technologies that best serve work and workers. NSF characterizes convergence as “the deep integration of knowledge, techniques, and expertise from multiple fields to form new and expanded frameworks” (NSF, 2017). To address the challenge of work and intelligent machines, it will be important to integrate perspectives and knowledge related to labor, incentives, motivation, cognition, machine learning, human learning, and systems design, among others. Our goal is to create conditions that bring researchers together to facilitate convergent research that illuminates the socio-technological landscape of the frontier of work and intelligent machines.