Confronting barriers to human-robot cooperation: balancing efficiency and risk in machine behavior.

Many technical and psychological challenges make it difficult to design machines that effectively cooperate with people. To better understand these challenges, we conducted a series of studies investigating human-human, robot-robot, and human-robot cooperation in a strategically rich resource-sharing scenario, which required players to balance efficiency, fairness, and risk. In these studies, both human-human and robot-robot dyads typically learned efficient and risky cooperative solutions when they could communicate. In the absence of communication, robot dyads still often learned the same efficient solution, but human dyads achieved a less efficient (less risky) form of cooperation. This difference in how people and machines treat risk appeared to discourage human-robot cooperation, as human-robot dyads frequently failed to cooperate without communication. These results indicate that machine behavior should better align with human behavior, promoting efficiency while simultaneously considering human tendencies toward risk and fairness.

Machine behaviour.

Machines powered by artificial intelligence increasingly mediate our social, cultural, economic and political interactions. Understanding the behaviour of artificial intelligence systems is essential to our ability to control their actions, reap their benefits and minimize their harms. Here we argue that this necessitates a broad scientific research agenda to study machine behaviour that incorporates and expands upon the discipline of computer science and includes insights from across the sciences. We first outline a set of questions that are fundamental to this emerging field and then explore the technical, legal and institutional constraints on the study of machine behaviour.

Cooperating with machines.

Since Alan Turing envisioned artificial intelligence, technical progress has often been measured by the ability to defeat humans in zero-sum encounters (e.g., Chess, Poker, or Go). Less attention has been given to scenarios in which human-machine cooperation is beneficial but non-trivial, such as scenarios in which human and machine preferences are neither fully aligned nor fully in conflict. Cooperation does not require sheer computational power, but instead is facilitated by intuition, cultural norms, emotions, signals, and pre-evolved dispositions. Here, we develop an algorithm that combines a state-of-the-art reinforcement-learning algorithm with mechanisms for signaling. We show that this algorithm can cooperate with people and other algorithms at levels that rival human cooperation in a variety of two-player repeated stochastic games. These results indicate that general human-machine cooperation is achievable using a non-trivial, but ultimately simple, set of algorithmic mechanisms.