Catastrophic Risk from Rapid Developments in Artificial Intelligence

Abstract

This article describes important possible scenarios in which rapid advances in artificial intelligence (AI) pose multiple risks, including to democracy and for inter-state conflict. In parallel with other countries, New Zealand needs policies to monitor, anticipate and mitigate global catastrophic and existential risks from advanced new technologies. A dedicated policy capacity could translate emerging research and policy options into the New Zealand context. It could also identify how New Zealand could best contribute to global solutions. It is desirable that the potential benefits of AI are realised, while the risks are also mitigated to the greatest extent possible.