Friedrich Hayek isn't a name we would generally associate with the rise of artificial intelligence, or AI. Born in Vienna in 1899, he was a statistician, an economist, and a philosopher of sorts; his name is generally associated with free-market, minimally regulated economics. He won a Nobel Prize in Economics, and that was back in 1974, when the Nobel still meant something, in the days before they started handing them out to people like Yasser Arafat, to political hacks like Paul Krugman, and to non-entities like Barack Obama.
Even so, Hayek's thinking and his theories on economics might lend some valuable insights into our current concerns with AI, with its growth, its increasing use in a wide range of applications, and how we might manage it — and how we might manage to control it. The key phrase comes from Hayek's Nobel acceptance speech: The pretense of knowledge.
It's illustrative to look at the people now agitating for controls over AI. City Journal's Adam D. Thierer has a few names.
Fears about artificial intelligence have gone mainstream, and an intense panic now grips the minds of the press, policymakers, and the public. A familiar cast of technocratic leaders—Barack Obama, Rahm Emanuel, Kamala Harris, Bernie Sanders, Elizabeth Warren, and even King Charles—have chimed in with calls for some sort of “AI pause” or other steps to give government major new powers to ensure proper “alignment” of AI models, algorithmic applications, and advanced computation.
It's more than a bit of a stretch to refer to most of these people as "technocrats." Most of them, especially Harris and Warren, have very little knowledge of either AI or economics, for that matter; Senator Warren in particular loves to finger-wag at us over economic issues while proving daily that her understanding of economics is on the level of a preschooler's understanding of quantum mechanics. But the shouting of these "technocrats" all has one thing in common: Calls for increasing government control. We are seeing, even now, the beginnings of that control; on Tuesday, the Trump administration, at the table with actual technocrats, signed a statement that is placing a few guardrails around AI.
But how effective will this be on a piece of technology that's still in its infancy? Hayek might caution us against getting too far out over our skis.
It’s a good moment to recall some timeless lessons from social thinker Friedrich Hayek. Today’s AI alignment debate contains echoes of the old economic calculation debate and the planner’s mentality that viewed society as a math problem to be solved. Hayek (1899–1992) spent his life demolishing that reasoning, while warning of the “fatal conceit” of elites who thought they could reorder entire economic systems and remake human nature through freedom-crushing, top-down interventions.
AI has one thing in common with economics; both are vast, complex systems beyond the understanding of most of us. But the question is this: What are we expecting AI to develop into? There's a possibility that some spontaneous order might arise that we didn't anticipate, which might well make our attempts to put guardrails around AI ultimately futile. In other words, AI systems may end up more like economies than we expected.
That's not to say that any AI will go full Skynet on us, and develop sentience and self-awareness. That still seems prohibitively unlikely. But like any new technology, it will likely find application and, more to the point, may adapt itself in ways that we didn't expect. Complex systems do that.
In his Nobel speech, Hayek warned us of the "pretense of knowledge."
It has, of course, to be readily admitted that the kind of theory which I regard as the true explanation of unemployment is a theory of somewhat limited content because it allows us to make only very general predictions of the kind of events which we must expect in a given situation. But the effects on policy of the more ambitious constructions have not been very fortunate and I confess that I prefer true but imperfect knowledge, even if it leaves much indetermined and unpredictable, to a pretense of exact knowledge that is likely to be false. The credit which the apparent conformity with recognized scientific standards can gain for seemingly simple but false theories may, as the present instance shows, have grave consequences.
Hayek was, here, writing about unemployment; AI was only the stuff of science fiction in 1974. But note the language: The true explanation of an economic metric, unemployment — a major concern in 1974 — is a "...theory of somewhat limited content because it allows us to make only very general prediction of the kind of events which we must expect in a given situation. But the effects on policy of the more ambitious constructions have not been very fortunate and I confess that I prefer true but imperfect knowledge, even if it leaves much indetermined and unpredictable, to a pretense of exact knowledge that is likely to be false."
Now, as a fun thought exercise, let's apply that to AI: We can, at present, only make very general predictions about this new technology; our early attempts to control and make use of AI are liable to be one of those "ambitious constructions" that are "not very fortunate." In other words, where AI is concerned, we may be operating only under a pretense of knowledge.
It's an interesting notion.
We should remember that AI is a technological tool, but it's also an economic tool. It's liable to change the way we do many things. And those economic impacts are going to include things we never expected. The adaptation of AI will cause some chaos in employment, in technology, in business, in military operations, in government; in fact we just don't yet know what all will come out of this. But, as Nietzsche points out, out of chaos comes order. AI will likely work its way into almost everything we do, in one form or another. And our attempts to put guardrails around it may prove futile. It's the interaction that's chaotic and unpredictable; not the AI itself so much as the use humans will put it to. And that, we will only learn, as Hayek reminds us, through experience. It will be a bottom-up, not a top-down, process, in other words — and that's probably the best way for us, in this new phase in our technological society, to handle it.
