AI Boosts Productivity: Hype, Reality or Mirage?

You are receiving this post/email because you are a patron/subscriber to Of Two Minds / Charles Hugh Smith. This is Musings Report 2024-1.

The consensus that AI will dramatically boost productivity and thus global prosperity in the coming decades has many enthusiastic analysts and few skeptics. For example:

The Coming AI Economic Revolution: Can Artificial Intelligence Reverse the Productivity Slowdown? (James Manyika and Michael Spence)

"In June 2023, a study of the economic potential of generative artificial intelligence estimated that the technology could add more than $4 trillion dollars annually to the global economy. This would be on top of the $11 trillion that nongenerative AI and other forms of automation could contribute. These are enormous numbers: by comparison, the entire German economy—the world’s fourth largest—is worth about $4 trillion. According to the study, produced by the McKinsey Global Institute, this astonishing impact will come largely from gains in productivity.

At least in the near term, such exuberant projections will likely outstrip reality. Numerous technological, process-related, and organizational hurdles, as well as industry dynamics, stand in the way of an AI-driven global economy. But just because the transformation may not be immediate does not mean the eventual effect will be small.

By the beginning of the next decade, the shift to AI could become a leading driver of global prosperity. The prospective gains to the world economy derive from the rapid advances in AI—now further expanded by generative AI, or AI that can create new content, and its potential applications in just about every aspect of human and economic activity. If these innovations can be harnessed, AI could reverse the long-term declines in productivity growth that many advanced economies now face."

Dave Friedman offered up some caveats about projecting trends based "pseudo-quantitative" analysis and reasoning:

AI & productivity: the economic effects.

"What I quickly realized was that forecasting is more art than science. Sure, there are trends, and these trends can be extrapolated. You can extrapolate linearly or non-linearly. You can tart up your extrapolations with things like moving averages and other pseudo-quantitative reasoning, but at the end of the day, forecasts are only as good as their comparison to actual results.

One of the more common AI-related forecasts being bruited about is that AI will usher in an era of sustained higher economic productivity. Higher economic productivity is good! It would solve many of our fiscal issues. So let’s think a bit about what this means. We may well be a few years away from a period of much higher economic productivity, and the second-order effects of that could be fairly profound."

Here’s a report from Goldman Sachs, titled Generative AI could raise global GDP by 7%:

"Breakthroughs in generative artificial intelligence have the potential to bring about sweeping changes to the global economy, according to Goldman Sachs Research. As tools using advances in natural language processing work their way into businesses and society, they could drive a 7% (or almost $7 trillion) increase in global GDP and lift productivity growth by 1.5 percentage points over a 10-year period.

“Despite significant uncertainty around the potential for generative AI, its ability to generate content that is indistinguishable from human-created output and to break down communication barriers between humans and machines reflects a major advancement with potentially large macroeconomic effects,” Goldman Sachs economists Joseph Briggs and Devesh Kodnani write in a report."

So Goldman has to tout the economic benefits of AI: there will be a boon in wealth creation over the coming decade so you (and your board) had better get on board with the AI hype train! And, of course, let our bankers lead the way.

Yes, I’m being cynical here. There is something to the claim that AI increases productivity. Just look at all the programmers who tell us how amazing Github Copilot is."

The consensus that AI will boost productivity is based on extrapolations of superficial abstractions that boil down to belief statements: AI will necessarily boost productivity because that's what it does, boost productivity. Well, yes, but exactly how, and by how much? How do we measure productivity? And what about the potential for AI to reduce productivity?  

Let's take some real-world examples.

Large Language Model (LLM) AI, a.k.a. generative AI, that can create new content based on natural language processing machine learning, is ultimately a  natural language interface with rather impressive limits on accuracy and what we might call self-awareness. That said, it's good at specific tasks such as composing a news report on a sports event, or generating a text summary of a podcast.

But in a world already awash in essentially free content, precisely how much productivity can be gained by generating limitless quantities of copycat content?  If the value proposition is attracting and keeping the attention of human consumers to collect data for target marketing, the super-abundance of content was never the value proposition, and therefore boosting the generation of more content to near-infinity isn't going to boost productivity in real-world terms.

Just creating near-infinite quantities of copycat content doesn't translate into producing more valuable output and productivity of either human workers or capital.

(Recall that LLMs are not free; they're voracious consumers of computer processing and memory, and electricity.)

As someone who creates, processes and works with content, it seems to me that generating an over-abundance of copycat content, in all media, will actually make the task of curating this near-infinity of near-zero-value content much more difficult. The enthusiasts will quickly say that AI will be tasked with curating the over-abundance of near-zero-value content, which is a massive drag on productivity created by AI itself.

This strikes me as an example of a metastasizing cancer being tasked with curing metastasizing cancer, or a diseased dragon eating its own tail. If we measure "productivity" as increasing sales and profits with less labor and capital, where is the productivity gain in overwhelming all media with a near-infinite spew of copycat content, the only purpose of which is to attract and lock in the attention of human consumers?

Another much-touted AI-based source of productivity gains is in customer service.