THE AI PARADOX

In this optimistic take, the seductive skills of AI systems are pitted against the complexity of human intelligence.

Will unchecked adoption of artificial intelligence overwhelm society? This book offers a reassuring counterpoint: AI’s limitations remain stark when compared with the richness of human intelligence. Dignum, professor of responsible artificial intelligence at Umeå University in Sweden, challenges the popular image of AI as an all-powerful brain. Organizing her chapters around a series of “paradoxes,” she argues: “The more AI can do, the more it reveals what makes human intelligence unique.” The author begins by examining conflicting definitions of machine intelligence. Weaving together research across fields is no simple feat, given the fragmented and varied AI landscape. Drawing on her experience in both corporate and academic settings, she observes that enthusiasm and anxiety abound, yet consensus is elusive—not only on what AI is, but also on the meaning of artificial general intelligence. Pushing back against doomsday scenarios, she contends that while skeptics such as Yoshua Bengio warn of catastrophic risks, AI systems lack humanlike motives, drives, and multidimensional intelligence. Instead of framing AI as an unstoppable force poised to subjugate humanity, Dignum encourages us to see it as an augmenting technology we can shape. The book is dense with thought-provoking intersections that could be expanded into another volume. Dignum situates AI within cultural, scientific, and corporate currents, though her analysis gives less attention to the crucial role of government funding in shaping the field, including the notorious “AI winters” when investment dried up after overhyped promises. Acknowledging that prominent AI pioneers have raised alarms, Dignum counters: “We should be more concerned about those who develop, own and deploy these systems.”

A nuanced, hopeful vision of a future with human intelligence amplified, not overwhelmed, by machine intelligence.