Computers as Complements, Not Substitutes for Humans

Key Insight

Information technology's rapid advancement leads to a widespread belief that computers will replace humans, evidenced by milestones like IBM's Deep Blue defeating Garry Kasparov in 1997, Watson beating Ken Jennings in Jeopardy! in 2011, and Google's self-driving cars raising concerns about chauffeur unemployment. Venture capitalists like Marc Andreessen assert 'software is eating the world', and questions like 'Will a machine replace you?' reflect growing anxiety. This fear mirrors past worries about globalization, where human workers in countries like Mexico were seen as substitutes for American labor, creating competition for jobs and resources.

However, computers fundamentally complement humans rather than substitute for them because their capabilities are vastly different. Humans possess intentionality, form plans, and make complex decisions, while struggling with enormous data processing. Conversely, computers excel at efficient data processing but fail at basic human judgments. For instance, a Google supercomputer, after scanning 10 million YouTube thumbnails, achieved only 75% accuracy in identifying a cat, a task an average four-year-old performs flawlessly, demonstrating a categorical difference between man and machine, not merely a difference in power.

On the demand side, computers do not compete for resources like people do; they merely require nominal electricity and have no desires for luxury goods. When technology is designed to solve problems, it acts as a hyperspecialized trading partner, offering efficiency gains without resource competition. Properly understood, technology is a means to escape competition in a globalizing world. The notion of 'strong AI' where computers entirely eclipse humans is considered a distant, 22nd-century concern, and indefinite fears should not prevent the development of technology that helps humans achieve previously unimaginable feats.