Only 13 Percent of Tech Leaders Report Big Returns from AI, Gartner Says
A Gartner survey of 2,002 technology leaders finds just 13 percent see major returns from AI, while 75 percent of CEOs still plan to raise tech spending.
A Gartner survey of 2,002 CIOs and other technology leaders suggests the AI payoff is still patchy: only 13 percent report major or transformative returns. The findings were presented at the Gartner IT Symposium in Orlando and reported by the Wall Street Journal and The Decoder.
Other numbers paint the same picture. According to the survey, 58 percent of CIOs have been asked to deliver cost savings through AI, yet 51 percent expect AI to raise the total operating cost of their tech stack. Meanwhile 44 percent of CEOs say they struggle to show stakeholders a financial return, and 75 percent still plan to increase technology spending. Gartner analyst Alicia Mullery describes a gap between the “innovation race” that vendors are running and the “outcomes race” that companies actually need to win. Companies doing well, she says, decide early which AI projects to continue and which to shut down, even ones already in production, and plan ahead for how agents will change job roles.
There is a counterpoint. Bloomberg CEO Vlad Kliatchko argues it is simply too early to measure the economic payoff, and Bloomberg is investing because it believes now is the right time. And a fair caveat on the data: the coverage is second-hand, the article does not say how “major or transformative” was defined, and it gives no details on sampling or industries. The respondents are technology leaders, not a cross-section of all companies, so the headline “companies” is looser than it sounds.
What’s behind it: this pattern is common with new technology. Buying a tool is fast, but changing how people work around it is slow. A company that adds a chatbot to an old process may save minutes; one that redesigns the process may save weeks. Surveys like this tend to capture the gap between those two groups. It is also worth remembering that analysts sell advice, so a “most firms are struggling” story is not neutral either.
What this means for you: if you are starting out, you do not need to feel behind. Most big organisations are still figuring it out. If you use AI at work, the lesson is to pick one concrete task, measure time or errors before and after, and drop what does not help. The question to ask is not “are we using AI?” but “what got better, by how much?”
Sources
Source: https://the-decoder.com/only-13-percent-of-companies-see-real-returns-from-ai-gartner-survey-finds/
Google Gives Every AI Agent Its Own Tiny, Locked-Down Computer
Google's Agent Compute Engine runs each AI agent in a separate hardware-isolated micro virtual machine that pauses when idle. A preview is planned for the coming months.
