DISCLAIMER: GoldInvestors.news is not a registered investment, legal or tax advisor or broker/dealer. All investment/financial opinions expressed by GoldInvestors.news are from the personal research and experience of the owner of the site and are intended as educational material. Although best efforts are made to ensure that all information is accurate and up to date, occasionally unintended errors and misprints may occur.

Generative AI is being promoted as an economic force capable of delivering enormous gains, even as critics warn about its negative consequences. Yet the productivity evidence most relevant to those sweeping efficiency claims has yet to confirm that the promised transformation is actually underway.

The caution comes with an important qualifier. AI development is moving quickly, and it was only last week that OpenAI released Dots, leaving plenty of time for the technology to produce measurable results.

For now, however, the AI boom is far more visible in corporate spending and soaring valuations than in broad productivity statistics. Apollo chief economist Torsten Sløk highlighted that disconnect in a blog post on Monday.

Sløk focused on the San Francisco Fed's “total factor productivity” index, commonly known as TFP. The measure tracks economic output derived from labor and capital, making it particularly useful for evaluating whether businesses are finding genuinely more efficient ways to operate.

Yahoo is a portfolio company of funds managed by affiliates of Apollo Global Management. That disclosure accompanies the analysis, which draws a clear distinction between investment fueled expansion and authentic productivity improvement.

The strength of TFP is that it holds steady both the number of hours worked and the amount of capital invested. It therefore can indicate whether companies are producing more without simply adding workers, machines, or other equipment.

That makes TFP a proxy for innovation and real efficiency rather than a scoreboard for spending. By this measure, the AI revolution remains conspicuously absent from the broader economic data.

Sløk's analysis found that TFP is currently sitting “slightly below zero with no sign of acceleration since the AI capex cycle began.” That reading stands in sharp contrast to the enthusiasm surrounding AI investments and valuations.

The gap could become increasingly important as corporate AI costs continue to rise. As those expenses expand, businesses will face greater pressure to justify adoption with measurable operating improvements rather than ambitious forecasts.

AI supporters often point to a different productivity statistic, output per hour from labor. That measure is running 2.5% above the average recorded since 2005, providing bulls with a seemingly persuasive number to support claims that AI is already delivering.

The problem is that stronger output per hour can reflect forces other than technological innovation. Sløk explained that capital deepening, which means giving workers better equipment, can increase their output without demonstrating that a company has discovered a fundamentally smarter process.

His example involved employees whose work depends on laptops. Providing everyone with a second monitor could allow them to accomplish more in an hour, but the improvement would result from additional equipment rather than a breakthrough in how the business operates.

That distinction matters when evaluating the 2.5% figure. Without holding labor and investment constant, observers may mistake ordinary capital deepening for a technology shock produced by AI.

TFP attempts to remove that ambiguity by accounting for both labor and capital. On that stricter test, current economic data does not show the acceleration that would confirm AI is already fulfilling its central productivity promise.

None of this proves that AI will fail to produce major economic gains. The technology remains in its infancy, while corporations may need considerable time to reorganize operations and capture the efficiencies that AI advocates expect.

Sløk also noted that a subzero TFP reading is not inherently unusual during the early stages of technological adoption. Electricity and information technology each required a decade or more before their effects became visible in the broader economic picture.

That history gives AI boosters a case for patience, but it does not provide evidence that the payoff has arrived. Investors can point to spending, valuations, portfolios, and output per hour, yet the broad efficiency measure remains unmoved.

The current picture is therefore more restrained than the market excitement surrounding AI might suggest. As Sløk put it, “the productivity payoff from AI remains a forecast rather than an observation.”

DISCLAIMER: GoldInvestors.news is not a registered investment, legal or tax advisor or broker/dealer. All investment/financial opinions expressed by GoldInvestors.news are from the personal research and experience of the owner of the site and are intended as educational material. Although best efforts are made to ensure that all information is accurate and up to date, occasionally unintended errors and misprints may occur.