Recently, the K-shaped economy has been the main topic of discussion, but the J-curve is also coming into play as people debate AI’s effect on productivity.
The curve means that general-purpose technologies such as AI don’t bring immediate benefits. Instead, heavy investment comes first, hiding early gains. Only after this initial decline does productivity truly take off, forming the J shape. However, for some, it’s not yet clear if this transformation is occurring.
Apollo Chief Economist Torsten Slok made a joke, referencing Robert Solow’s well-known quote about the PC revolution. Slok also said that employment, productivity, and inflation figures still don’t show signs of the new technology. At the same time, profit margins and earnings projections for S&P 500 companies outside the “Magnificent 7” also don’t show evidence of AI in action.
He wrote, “Maybe there’s a J-curve effect for AI, where it takes time for AI to appear in macro data. Maybe not.”
But in an article titled ‘The AI Productivity Take-off is Finally Visible,’ economist Erik Brynjolfsson cited the latest jobs report from the Bureau of Labor Statistics as proof that ‘the fog may finally be lifting.’
Wednesday’s report adjusted the 2025 job gain figure to only 181,000, down from the initial 584,000 and 2024’s 1.46 million gain.
Since the economy kept growing steadily last year while adding very few workers, with fourth-quarter GDP estimated at 3.7% growth, that indicates a surge in productivity.
Brynjolfsson stated that his own analysis indicates U.S. productivity rose about 2.7% in 2025—nearly twice the 1.4% annual average of the past decade.
He said, “The updated 2025 U.S. data shows we’re now moving from this investment phase to a harvest phase where those earlier efforts start to show as measurable output.”
Brynjolfsson, who is director of Stanford University’s Digital Economy Lab and has been studying AI since before ChatGPT stunned the world, published a study last year that showed AI was boosting productivity, especially among those aged 22 to 25 in highly AI-exposed professions.
He warned that several more periods of sustained growth are needed to confirm a long-term productivity trend, noting that geopolitical or monetary problems could counteract progress.
But while many businesses are still using AI minimally, Brynjolfsson said he found ‘a small group of power users’ who are using AI agents to automate end-to-end work processes, finishing tasks in hours instead of weeks.
He wrote in the FT, “We’re moving from an era of AI experimentation to one of structural utility. We now need to focus on understanding its exact mechanisms. The productivity revival isn’t just a sign of AI’s power; it’s a wake-up call to focus on the upcoming economic transformation.”
When looking at the information and communication technology (ICT) sectors, others also see clear signs that AI is increasing productivity.
Stephen Brown, chief deputy North America economist at Capital Economics, said in a note earlier this month that ICT output increased in the third quarter even though employment decreased.
He also said that while earlier payroll cuts were probably because of overhiring during the pandemic, reductions have continued even as ICT sectors have flourished.
Brown stated, “All this means AI is making a significant contribution to productivity growth.”