An HFS Research study conducted with Rackspace found that surveyed companies reported AI productivity gains more often than revenue growth, while only 33% said their main AI program met or exceeded its business case.
Rackspace said the survey covered 502 senior enterprise AI decision-makers in August and September 2026. Respondents worked at companies with annual revenue of at least $100 million across eight industries and four regions. All their companies were running AI pilots or more advanced programs, and 83% had AI in production or beyond.
HFS reported that 58% of respondents cited a more productive workforce and 42% reported lower costs. Improved customer experience was cited by 28%, while 21% reported a measurable increase in revenue.
The study also found significant AI governance gaps among 96% of respondents. Seventy-eight percent viewed governance failures as a bigger barrier to expanding AI use than technical failures. The most frequently cited shortcoming was oversight of autonomous AI after deployment (55%); 44% said their organization lacked a single, consistent way to operate AI in production.
Business-case results varied across deployment stages. The share saying a program met or exceeded its case was 37% at companies using AI in one function, 28% at those using it across several functions and 56% where AI was part of standard operations. HFS cautioned that these figures describe separate groups and do not establish that operational maturity drives stronger returns.
HFS traces weak results to unclear ownership, poor repeatability and limited visibility among the people and providers managing AI. The report compares the challenge with cloud and outsourcing, where implementation work can end while the job of operating a live system continues. HFS Research founder and CEO Phil Fersht said, “Enterprise AI has an accountability problem that nobody in the industry wants to admit.”
HFS recommends assigning an operator to track AI systems’ performance, costs and governance once they are in use. That responsibility could sit with an internal team, providers under a shared management structure or a managed-services firm; the company would retain control of its business priorities, policies, decisions and legal responsibilities.
In the survey, 61% preferred one accountable partner across models, platforms and infrastructure. Another 73% said they would pay more than raw infrastructure costs for a fully managed service with governance accountability; these responses measure stated preferences, not purchases. HFS said providers could earn a premium by demonstrating clear responsibilities and measurable outcomes. Rackspace’s partnership on the study and its description of itself as an enterprise AI-stack operator point to a commercial interest in the model HFS recommends.
The results describe responses from AI-active firms, not the prevalence of AI adoption across all companies.
