According to tech entrepreneur Azeem Azhar, two-thirds of approximately 160 IT vice presidents surveyed at an event reported seeing measurable results from artificial intelligence implementations within their organizations. However, when asked if those results were significant enough to interrupt their CEO’s vacation, only about eight executives kept their hands raised, highlighting a slow pace of high-impact enterprise ROI.

While trust and enterprise AI budgets continue to grow, many companies are shifting away from expensive frontier models toward open-weight alternatives to control costs. This shift poses potential risks for major AI labs, as growing adoption may not translate into sufficient revenue to sustain massive infrastructure investments.

Why it matters

  • Enterprise AI adoption is expanding, but high-value ROI remains rare, driving interest toward lower-cost open-weight models.

  • The gap between infrastructure spend and direct enterprise revenue poses a strategic risk for frontier model developers.

Source: the-decoder.com