A computer’s victory over a human go master this past March reminds us of the pending “singularity” — the rapidly approaching moment in time when artificial intelligence overtakes human intelligence. Machines will learn, and we won’t be their teachers. Are we prepared for it? Can we prepare for it?
We’d better. Many futurists declare it inevitable, probably within a generation, maybe less. Shukan Shincho magazine discusses some hypothetical implications in its Aug. 25 edition. Even the least of them are shocking. For example, in 2045 a computer with the combined intellectual power of the entire human race would cost $100. In short, it’ll be no big deal. What will be a big deal? Should we shudder at the thought, or rejoice?
Francis Bacon (1561-1626) is generally acknowledged as the grandfather of modern science. “Nature, to be commanded, must be obeyed,” he wrote. His fictional “New Atlantis” was a utopia ruled by scientists who, having admitted their ignorance and purged themselves of illusory knowledge, experimented, observed and slowly built up from scratch an ever-expanding store of “true” knowledge — armed with which they “commanded” nature to outgrow her destructive caprices and ease mankind’s lot.
It’s his vision we’ve lived by, more or less, ever since. We’re here. It’s not exactly paradise, but unquestionably, thanks to theoretical and applied science, our passage through life is smoother, more comfortable and longer than it’s ever been before.
It would be interesting to ask Bacon what he thinks of it all. From wind power through steam power to nuclear power is an awesome voyage. We command, nature obeys. Sometimes we command unwisely and pay the price. Sometimes nature rebels, and that’s costly too, but she’s soon brought to heel and discipline is restored. The march of progress is uneven but unstoppable.
Now artificial intelligence looms. Could Bacon have foreseen it? One imagines him imagining nuclear power more easily. Its advent was amazing but not quite, perhaps, a “singularity.” The term was originally coined by cosmologists to describe the big bang that hypothetically produced the universe. First there was nothing, then there was something. The border between those two states is the singularity.
The use of the same word in the context of artificial intelligence suggests another border, scarcely less consequential. Matter converted into energy should have qualified, but former U.S. President Harry Truman, speaking of his decision to atom-bomb Japan, said, “It was just the same as getting a bigger gun than the other fellow.” If he’d lived to see Google DeepMind’s AlphaGo computer program humble go master Lee Sedol, would he have said, “It’s just a slicker machine”? Some imaginations are incurably pedestrian.
What Truman failed to see with foresight, we see with hindsight. He launched a new age and didn’t even know it. Maybe AlphaGo is our Truman, with very little more idea of what its triumph represents, long-term, than the president had as he issued his fateful command.
The border crossed, the singularity traversed, where will we humans stand? On very unsteady ground, judging from Shukan Shincho’s jolting headline, though the tenor of the report, oddly enough, is benign, even positive — business as usual, only more so. “A bigger gun.” The headline reads: “Twenty years down the road, all of us who are not (technological) geniuses will be unemployed.”
That’s actually a bit misleading. Menials, too, will have at least some of their work cut out for them. Take cleaning, for instance. There’s Roomba, the robot vacuum cleaner. Whether its ability to navigate a room full of obstacles and distinguish dirt from non-dirt qualifies as intelligence hinges on a definition. Either way, it’s marvelously capable — and yet limited: It can’t (for now) do desks, walls or ceilings.
That seems very fragile security for human cleaners of desks, walls and ceilings, but Kobe University professor Takuya Matsuda cites it to show that some forms of menial employment will remain. Post-singularity Japan, he forecasts, will retain an employment rate, geniuses and menials combined, of roughly 10 percent.
And the remaining 90 percent? What of them?
A concept that has been around for a while (its roots antedate Bacon) is known as basic income. It would give everyone — employed and unemployed, rich and poor — a guaranteed minimum income, not as a gift, handout or bribe but as a matter of entitlement. Centuries of discussion on the subject have been intellectually stimulating but largely inconsequential in terms of practice. Post-singularity, suggests Riken researcher Koichi Takahashi, that’s how most of us will draw our pay.
Fantastic. There’d be no guilt because unemployed recipients would constitute a vast majority. And if the $100 price tag, mentioned above, on an omniscient, probably prescient, possibly even imaginative computer is any indication, devices to live our lives for us would be readily affordable. Food and drink might be less so, but a basic income would at least keep us from starving, and any non-genius who wanted to work, whether to stave off boredom, slake ambition or live luxuriously, would be perfectly free to start a business. There would surely be some human needs that intelligent machines would miss — the equivalent of the walls and ceilings that Roomba can’t clean. Entrepreneurs would find them and fill them for profit, while the less enterprising or less restless would be … eternal consumers? Why not?
New Atlantis, or Brave New World? If Shukan Shincho is worried, it doesn’t show. Its one regret, apparently, is that Japan lags “20 years behind” leading countries in the field. The U.S. churns out 57 percent of the world’s theses on AI and IT, the EU 18 percent and China 8 percent. Japan contributes a mere 2 percent.
The culprit here, says University of Tokyo engineering researcher Yutaka Matsuo, is Japan’s retreat from the hands-on manufacturing at which it once excelled. Let it recover its edge in that field, Matsuo says, and it will become a leading producer of a new generation of “machines with eyes.”
How will they use those eyes? To look for new ways to serve us? Or to keep us, their intellectual inferiors, in line as we fall ever farther behind them, languishing as they flourish, waning as they wax?
Michael Hoffman is the author of “In the Land of the Kami: A Journey into the Hearts of Japan.”
In a time of both misinformation and too much information, quality journalism is more crucial than ever.
By subscribing, you can help us get the story right.