Superintelligence and Natural Resources: Morality and Technology in a Brave New World

A growing majority of artificial intelligence (AI) scientists and engineers believe human-level intelligence will be attained before the dawn of the 22nd century, while a large number believe this will happen much sooner. At this point, the elevation to “superintelligence,” or systems that have superhuman levels of general intelligence, would likely proceed very rapidly. With this in mind, this commentary has three goals: (1) to provide an overview of how a leading scholar, Nick Bostrom, conceives of superintelligence, with its potential and risks, (2) to bring attention to a burgeoning form of common pool resource extraction, space mining, that may benefit from superintelligence management, and (3) to suggest that environmental social scientists should play a key role in programming seed AI, infusing it not only with our existing collective knowledge, but also with just values.