Enterprises rush to adopt AI but struggle to measure its value

New research shows that enterprises are going beyond AI experimentation and into large-scale production but that return on investment is taking a back seat in the process.

The report from Domino Data Lab shows that while 88 percent of organizations report improved ability to move AI from experimentation to production, nearly 60 percent expect less than 50 percent ROI in the rapidly changing areas of machine learning (60 percent) or Gen AI (57 percent).

"It's clear that the AI hype cycle has peaked. Enterprises are done experimenting, and they're racing to scale, but the payoff is still out of reach," says Thomas Robinson, chief operating officer at Domino. "Our data validates what leaders know about and followers are learning: aligned leadership, flexible infrastructure, and integrated governance drive a straight line to AI business impact. These aren't technical upgrades -- they're business imperatives that will define who sees real returns from AI."

Integrated AI/ML governance is the top capability required for AI execution by enterprises, (66 percent) followed closely by tooling (49 percent), and hybrid cloud AI infrastructure (44 percent). More than half (51 percent) are focused on securing and structuring data pipelines to make their infrastructure AI-ready today and tomorrow.

Infrastructure to support AI across global clouds and on-premises environments is becoming increasingly important, with 48 percent of leaders believing hybrid cloud is a critical capability. Among organizations that have not yet formalized all their AI initiatives, 42 percent still identify hybrid infrastructure as a top priority.

Cost remains a critical concern. From acquiring AI talent (63 percent), to managing unexpected project expenses (61 percent), to challenges in leveraging existing infrastructure (61 percent), financial strain continues to hinder the realization of AI's full value. The price tag for developing and operating AI (60 percent) was also highlighted as a major barrier by respondents.

Looked at by industry, life sciences enterprises have the highest adoption rate of cutting-edge AI techniques, with 98 percent using GenAI and 95 percent using agentic AI. Finance organizations though are the most mature, with more than half (51 percent) of respondents saying that their organization has formally designated AI leadership roles and 66 percent of respondents identifying integrated AI/ML governance as their top execution priority.

You can get the full report from the Domino site.

Image credit: Thawatchai Chawong/Dreamstime.com