[DigitalToday reporter Chi-gyu Hwang (황치규)] There are emerging signs that companies selling AI applications are finding it more difficult to sell than before, and they’re gaining attention.
A recent Wall Street Journal report said AI software companies say they are taking longer to make purchasing decisions than last year as companies shift to cautious mode.
Last year, a combination of board-level directives, fear of missing out (FOMO), and aggressive campaigns by technology companies led companies to invest aggressively in AI. Market research firm Gartner estimates that $1.249 trillion was spent on software alone last year.
These days, companies seem to be more cautious about purchasing AI software than they were last year. The Journal cited AI-based customer service startup Legal as an example, reporting that last year deals could close in 60 to 90 days, but now it typically takes 180 days to complete a purchase.
“There was a time when early adopters embraced innovative technologies rapidly, but recently that pace has slowed significantly,” said Regal CEO Alex Levin. Craig Ross, a vice president at Gartner, agreed: “Everyone is becoming more cautious. I think they’re starting to recognize the reality.”
The analysis shows that companies have become cautious, in part because they have realized that rapid adoption is not always better. The Journal reports that early adopters that rushed to pilot and fully deploy AI last year often hit a wall and learned costly lessons along the way.
A National Bureau of Economic Research (NBER) survey of 6,000 executives, including CEOs and chief financial officers (CFOs), from companies in the United States, United Kingdom, Germany, and Australia also found that the productivity gains from AI remain modest.
Fortune reported that two-thirds of survey respondents said they use AI, but only spend 1.5 hours a week using it. 25% of respondents said they do not use AI at all in their workplace. It also said that 90% of responding companies said AI had no impact on hiring or productivity over the past three years.
Business expectations regarding the impact of AI on business and the economy remain high. In an NBER study, executives predict that AI will increase productivity by 1.4 percent and output by 0.8 percent over the next three years. Some believe that AI-driven productivity may follow a “J-curve,” with an initial slowdown before a sharp rise.
However, a gap still remains between expectations and reality. ManpowerGroup’s 2026 Global Talent Barometer, based on a survey of nearly 14,000 people in 19 countries, found that regular use of AI increased by 13% in 2025, while confidence in the technology’s usefulness decreased by 18%, according to Fortune magazine.
Ross said this wasn’t because the technology itself didn’t work, but because proper safeguards weren’t in place or companies didn’t fully understand the realities of the business processes they were trying to automate. “Of particular importance,” he added, “is that they found it difficult to measure financial performance. Even the performance that could be measured was not particularly impressive.”
Gartner predicts that enterprise software spending will increase 14.7% year over year to approximately $1.434 trillion this year. However, purchasing styles seem to be becoming more demanding. Ross reportedly said that as companies mature enough to understand potential roadblocks, they are taking longer evaluation periods and taking a more critical approach to the solutions they are adopting.
