Adversarial risks in social experiments with new technologies

Studies that approach the deployment of new technologies as social experiments have mostly focused on unintentional effects, notably safety. We argue for the inclusion of adversarial risks or security aspects that are the result of intentional, strategic behavior of actors, who aim at using the technology for their own purposes. Examples include cybercrime and terrorist use of materials obtained from nuclear facilities. Our general question is how the approach of new technologies as social experiments can be adapted or extended to account for such aspects. To this end, we re-interpret adversarial risks in social experiments in terms of actor-network theory, highlighting how intentional actions involving new technologies are different from unintentional events, by focusing on how actors form new networks with new technologies (composition), thus enabling new programs of action (translation). We also derive a typology of ways in which intentional (mis)use of new technologies creates risks, using the Bitcoin distributed currency as an example. Both contribute to conceptualizing adversarial risks in social experiments, as well as laying bare options for learning about adversarial risks during such experiments.