History defines the present with a clarity that is unavailable by any other measure. This became particularly obvious in “The Economics of Superintelligence” and “Eureka All Day Long”, two articles in the July 26th edition of The Economist, in which the digital pundits of Silicon Valley are trying to prepare us for the possible economic impact of artificial intelligence on our modern world. But, to do this, the author or authors of the articles (unidentified) have provided us with some pertinent economic history of the West to provide an illuminating context.

From the fall of the Roman Empire to 1700, “global economic output had expanded by 0.1% per year, a rate at which it took nearly a millennium to double.” Expressed differently, the world’s economy grew by about 8% per century, and closely followed the growth in population.

As for inflation, it remained ostensibly flat, at least in Europe, for the roughly 500 years between about 1000 to 1500, which meant the price of bread, wine, houses and labour never increased. The first significant increase in inflation occurred in the 17th century as gold and silver began to pour into Europe from its pillaging of the New World. Today, in our modern reckoning, 2.0% inflation per year is considered acceptable—even economically beneficial.

As for economic growth, it “quintupled to 0.5% per year between 1700 and 1820 with the advent of the early Industrial Revolution. During the next 300 years, compared to the 8% increase per century, economic growth “averaged 350%” per century. Affluence grew, mortality plummeted, populations exploded, new inventions multiplied, and the combination of these factors accelerated the whole process. “By the end of the 19th century, [annual economic growth] had reached 1.9%. In the 20th century it averaged 2.8% per year, a rate at which production doubles every 25 years.”

“If the evangelists of Silicon Valley are to be believed,” The Economist notes, “this bang is about to get bigger. They maintain that artificial general intelligence (AGI), capable of outperforming most people at most desk jobs, will soon lift annual GDP growth to 20-30% a year, or more. That may sound preposterous, but for most of human history, they point out, so was the idea that the economy would grow at all.” Another think tank, Epoch AI, estimates that once AI can replace 30% of human jobs, annual economic growth will exceed 20%.

All this visionary speculation doesn’t seem to consider what all the displaced people will be doing, how they will earn money, and how they will participate in the economy without a radical redistribution of wealth. No widely distributed wealth means no consumers, which means no economy.

Artificial intelligence can design better systems, solve difficult problems, and technically produce better products, but it is also energy intensive—media corporations are already scrambling to produce enough electricity needed to power their AI projects.

But, perhaps the most salient question pertains to where all this economic activity is going to occur. We are already experiencing the limits to growth on a finite planet. Resources are being stressed, natural ecologies are collapsing, greenhouse gases are not being appropriately limited, global temperatures are rising, food production is being threatened, the intensity of droughts and storms is increasing, menaced populations from climate change are in mass migration, and digital dreamers are imaging a Brave New World of AI.

AI is being used to solve complex problems that have avoided human ingenuity, so it could be used to propose solutions to our global environmental problems. But would any proposed solutions be implemented? We don’t need AI to tell us that we must curtail and eliminate greenhouse gas emissions to avoid an existentially sobering outcome for the entire planet. That simple solution is not being adequately employed. AI’s prospective solutions could design more efficient devices to produce clean energy, collect and store atmospheric carbon dioxide, or even adjust genes so that food crops are higher yielding and more weather tolerant. The obstacle, however, is not technology but our human psychology.

This, however, is the technological trap of hypothetical solutions. In a world that is racing toward environmental Armageddon, we don’t have the time to doddle while the planet burns. We don’t have time to wait while AI reconfigures our entire economic system. And we certainly don’t have the time to waste while we wait for a change in our human character. We know what to do, so why don’t we just do it before it’s too late?

As history attests, we have come some distance in the last few centuries. The Industrial Revolution conferred incredible comfort, riches and power on us. But it also mass produced ecological disasters. We seem to have reached a crossroads that may be even more significant and pivotal than anything that has happened in our history as a species. If we recognize this as being so, then we must act with the urgency and the resolve that the situation requires.

Ray Grigg for Sierra Quadra