The Myth of Artificial Intelligence: Why Computers Can't Think the Way We Do

The book "The Myth of AI" explores the gap between the actual science underlying AI and the dramatic claims being made for it, arguing that AI works on inductive reasoning and cannot replicate human intuition and creativity. It warns that we are making two mistakes: defining human intelligence down and overestimating what AI is likely to achieve.

Format: Paperback / softback
Length: 320 pages
Publication date: 28 October 2022
Publisher: Harvard University Press

The vast chasm between the actual science underlying artificial intelligence (AI) and the grandiose claims being made for it is exposed in Erik Larson's groundbreaking book. The book delves into the realm of AI, illuminating why this misguided association between AI and human intelligence is holding back progress. Larson, a computer scientist specializing in natural language processing, takes readers on a comprehensive tour of the AI landscape, revealing the underlying mechanisms and the limitations that hinder the development of true artificial intelligence.

AI operates on inductive reasoning, analyzing vast data sets to make predictions. However, humans possess a unique ability to make conjectures, drawing from context and experience. The challenge lies in programming intuitive reasoning, which forms the foundation of common sense. Futurists predict that AI will eventually surpass the cognitive abilities of the most gifted minds, but Larson cautions that we are still far from achieving true superintelligence.

One of Larson's primary concerns is the tendency to define human intelligence down, as we strive to replicate its capabilities in AI. This reductionist approach overlooks the profound differences between human and machine cognition. While AI can process vast amounts of data and make accurate predictions, it lacks the inherent creativity, empathy, and critical thinking skills that define human intelligence.

Another concern raised by Larson is learned passivity. As we rely more on AI to solve problems, we tend to assume that it will always find a solution. This complacency undermines our human ingenuity and creativity, as we become less inclined to solve problems ourselves. We need to cultivate a mindset that recognizes the value of human intelligence and encourages us to explore new solutions and innovate.

In conclusion, Erik Larson's book serves as a timely warning against the overestimation of AI's capabilities and the underestimation of human intelligence. By illuminating the gap between the two, the book encourages us to reevaluate our approach to AI and recognize the importance of nurturing and enhancing our human capabilities. Only by combining the strengths of both human and machine intelligence can we achieve true progress and realize the full potential of AI.

Dimension: 210 x 140 (mm)
ISBN-13: 9780674278660