A New AI Social ContractBy
Most of our attention in the emerging artificial intelligence (AI) world order has been on who will win the race for supremacy and how machine intelligence will be implemented in our daily lives. But the other long-range effects of AI will have far greater economic, political, and social ramifications that have yet to capture much notice.
In his latest book, AI Superpowers (Houghton Mifflin Harcourt), Kai-Fu Lee has created a seminal text that considers the history of AI, the present international competition, and various insights into the complicated future humanity is creating for itself with artificially intelligent devices, assistants, and human-sized androids. Lee has 30 years of experience in AI research and development at Apple and SGI, was once president of Google Greater China, and was founding director of Microsoft Research Asia. His vision of a future with AI is illuminated by his considerable experience.
THE REAL AI CRISIS
As the current debate between the utopians and the dystopians paints two drastically different futures if and when we achieve AGI (artificial general intelligence), Lee describes a serious problem that’s grounded in data and current realities, not conjecture: “This, I believe, is the real underlying threat posed by artificial intelligence: tremendous social disorder and political collapse stemming from widespread unemployment and gaping inequality.”
The source of the collapse is tied to AI’s disruption in the workforce. “Within 15 years,” he writes, “artificial intelligence will technically be able to replace around 40-50% of jobs in the U.S.” To estimate the possible upheaval that would cause, we have an historical reference point when the United States recorded its highest unemployment figure. That was the Great Depression. In 1933, 24.9% of the workforce was unemployed. But that number is only half of what Lee foresees as possible in our future.
It’s true that we have adjusted to other kinds of upheaval since the first Industrial Revolution, but Lee explains that the AI revolution is of a different magnitude. “Make no mistake: this is not just the normal churn of capitalism’s creative destruction…. The free market is supposed to be self-correcting, but these self-correcting mechanisms break down in an economy driven by artificial intelligence. Low-cost labor will provide no edge over machines, and data-driven monopolies are forever self-reinforcing.”
In the past, the preponderance of material goods and the limits of geography helped control monopolies. Distribution would involve a network of others who would share the manufacturer’s profits. With digital products, “instead of a dispersion of industry profits across different companies and regions, we will begin to see greater and greater concentration of these astronomical sums in the hands of a few, all while unemployment lines grow longer.”
Lee feels this new global distribution of wealth will not only be more unequal but hopelessly so, with the further risk that AI will create a new caste system, “one that divides the population into the AI elite and what historian Yuval N. Hatari has crudely called the ‘useless class,’ people who can never generate enough economic value to support themselves.”
Lee also discusses another human dimension of the AI revolution—the psychological loss of one’s purpose. “We have been conditioned to derive our sense of self-worth from the act of daily work. The rise of AI will challenge these values and threatens to undercut that sense of life-purpose in a vanishingly short window of time.”
The methods for facing this historically unique challenge might require reconstructing our economies and rewriting our social contracts.
Today, the three most suggested fixes are only “technical fixes,” Lee explains, “tweaks to policy and business models to ease the transition.” They are called the three R’s and are the most commonly proposed by Silicon Valley: reduce work hours, retrain workers, and redistribute the wealth.
After explaining how the first two are important but incapable of dealing long-term with the changes, Lee defines the most popular redistributive proposal—universal basic income (UBI). With UBI, “every citizen (or every adult) in a country receives a regular income stipend from the government—no strings attached.” An alternate proposal, guaranteed minimum income (GMI) is an “income floor” that would be given only to the poor.
Lee takes a larger step forward to something he calls human-AI symbiosis. “I propose,” he writes, “we explore the creation not of a UBI but of what I call a social investment stipend. A decent government salary given to those who invest their time and energy in those activities that promote a kind, compassionate, and creative society.” Just as we now reward economically productive activities, we will reward socially beneficial activities. The three pillars of this new social contract would be care work, community service, and education.
Fifteen years isn’t a very long lead time to prepare for changes of this magnitude, and serious planning is needed now.