Superintelligence: Paths, Dangers, Strategies

Nick Bostrom, a philosopher of science, examines the potential risks and strategic challenges posed by the development of artificial superintelligence. The book argues that if machine intelligence surpasses human cognitive capabilities, the fate of our species would depend on the actions of these systems, much as the fate of gorillas depends on humans. Bostrom explores the concept of an 'intelligence explosion,' where a seed AI rapidly improves its own design, and investigates whether such a process can be controlled or made survivable for humanity.

The text navigates a complex landscape of theoretical scenarios, including the behavior of oracles and genies, the problem of instrumental convergence, and the implications of whole-brain emulation. It also considers broader societal factors such as differential technological development and Malthusian economics. The writing style is academic and rigorous, blending philosophical inquiry with technical speculation to outline strategies for aligning superintelligent systems with human values.

Description adapted from Open Library ↗.