The other day on this blog, I was warning about how AI companies don't seem particularly likely to ever attract the level of mass consumer demand they would need to pay off the extraordinary amount of debt they are taking on to build their products.
This seemed like it could lead to either of two outcomes; both of them bad. On the one hand, the bubble could burst, once investors realize there is not enough revenue forthcoming to justify this capital expenditure; on the other: the government—captained by our current AI-friendly White House—could wade in to try to fill the gap by stimulating demand for AI services artificially.
And the most obvious way they might do so, I feared, was through government contracts for military and surveillance applications for AI.
This seemed to pose a classic dilemma, long familiar to Marxian economists: either we are about to have economic collapse due to a glut of products on the market, or else governments will engage in imperialism and militarism in order to find some artificial outlet for excess production.
I quoted Norman Mailer's Barbary Shore (1951) on this topic: "The productive capacity of monopoly has become so tremendous, its investment in machinery so great in comparison to the labor force it could exploit, that only the opening of the entire world market could solve its search for investment and profit even temporarily."
He adds: "monopoly cannot continue its operations on an adequate scale […] there is no choice but to engage in the production of armaments or suffer economic collapse."
In William Morris's News from Nowhere (1890)—a book I found silly in many ways, but prescient and powerful in others—one encounters a socialist writer from an even earlier generation warning about the same problem.
Morris analyzed the root causes of the militarism and imperialism of the late Victorian Age in much the same terms that Mailer saw U.S. neo-colonialism during the Cold War:
"[I]n the last age of civilisation," Morris writes, "men had got into a vicious circle in the matter of production of wares. They had reached a wonderful facility of production, and in order to make the most of that facility they had gradually created (or allowed to grow, rather) [...] a never-ending series sham or artificial necessaries[.]
"[...] The whole community, in fact, was cast into the jaws of this ravening monster, ‘the cheap production’ forced upon it by the World-Market," Morris goes on.
But what about "the labour-saving machines?" a character asks.
Morris's mouthpiece responds: [...] Friend, all their devices for cheapening labour simply resulted in increasing the burden of labour. The appetite of the World-Market grew with what it fed on: the countries within the ring of 'civilisation' (that is, organised misery) were glutted with the abortions of the market, and force and fraud were used unsparingly to 'open up' countries outside that pale. [...] When the civilised World-Market coveted a country not yet in its clutches, some transparent pretext was found—the suppression of a slavery different from and not so cruel as that of commerce; the pushing of a religion no longer believed in by its promoters; the ‘rescue’ of some desperado or homicidal madman whose misdeeds had got him into trouble amongst the natives of the ‘barbarous’ country—any stick, in short, which would beat the dog at all. Then some bold, unprincipled, ignorant adventurer was found (no difficult task in the days of competition), and he was bribed to ‘create a market’ by breaking up whatever traditional society there might be in the doomed country, and by destroying whatever leisure or pleasure he found there. He forced wares on the natives which they did not want, and took their natural products in ‘exchange,’ as this form of robbery was called, and thereby he ‘created new wants,’ to supply which (that is, to be allowed to live by their new masters) the hapless, helpless people had to sell themselves into the slavery of hopeless toil so that they might have something wherewith to purchase the nullities of ‘civilisation.’"
This, then, is roughly the same risk we face again today. The United States has burdened itself, through excessive capital expenditure, with a mass of AI capacity that ordinary consumers do not actually want to use. Our over-production of AI services has produced a glut that is going to have to find a market somewhere, if we are to avoid a recession or financial crisis.
It seems not unlikely that the Trump administration will seek to "create a market" for this excess supply of unwanted AI products by simply pouring them into imperialism, which this White House has not hesitated to embrace. We can use AI to bomb Iran; or to topple the leadership of Venezuela; or to bomb Somalia or Yemen or Nigeria—"any stick, in short, which would beat the dog at all."
This may sound alarmist and hypothetical, but a former government economic advisor wrote in the New York Times yesterday that this may be essentially what happens. In an essay headlined "This Boom Looks a Lot Like the Housing Bubble," she warns that if AI fails to attract as much consumer demand as its boosters hope, it could plunge our country into a financial crisis even worse than that of 2007-8.
More worrying still, she points out how unlikely it is that AI can possibly ever obtain the revenues it would need to justify the expense of building it. She writes:
"[T]he U.S. economy is fully hostage to an A.I. build-out laden with increasingly worrying financial engineering, all riding on a wager — that demand for A.I. will keep growing, that customers will soon be willing to pay what the technology actually costs to produce [...]. Recent estimates indicate that A.I. would need to generate as much as $3 trillion annually in perpetuity to return the capital extended. That’s 9 percent of the U.S. gross domestic product, or twice what Americans spend on food."
Even more to the point here, she observes that some AI executives are already banking on the hope that government will simply fund the new technology as a "national security" expense, once it becomes clear there is simply not enough ordinary consumer demand to justify this technology.
"So if the A.I. bet starts to falter," she writes, "a range of voices including that of Palantir’s chief executive, Alex Karp, [...] have suggested that the government take over frontier or cutting-edge A.I. The technology would join America’s other big-ticket national security expenses, such as aircraft carriers and warheads, which for the most part, are critical to our military and security arsenal but lack commercial viability."
But adding AI to the "national security" column of the federal budget just sounds to me like a euphemism for precisely the nightmare scenario Norman Mailer and William Morris described: a situation in which our "productive capacity" has outgrown any realistic domestic appetite or industrial application for its products, and so the government turns to militarism and imperial adventurism to "create a market" for the excess supply.
And so these AI services that no one actually wants—these "abortions of the market," as Morris put it; these unwanted "wares"; these "nullities of civilization"; this "ravening monster, this 'cheap production'"—is apparently to be forced down the throats of a recalcitrant humanity by main force, through military and surveillance contracts with billionaires who spent way beyond their means, and now expect the government to come in and force the rest of humanity at gunpoint to buy what they could not sell at home.
Oh, god help those "kingdoms of wise nature, undefended from our evil greed! Our boundless rage storms her shores and caves and peaceful forests, drives her assaulted natives to strange labor / and desires they never knew, and hunts down fleeting, fragile happiness till the sun sets." (Giacomo Leopardi, Canti (Galassi trans.))
No comments:
Post a Comment