A global study by IBM, an American multinational technology and consulting company and Oxford Economics, a global economic forecasting and advisory firm, reports that only 18% of Chief AI Officers (CAIOs) have full control over AI investment decisions, despite being positioned to see where the technology creates value.
Based on a survey of 1,667 senior AI executives across 33 geographies and 20 industries conducted between April and June 2026, the findings highlight a gap between accountability and authority: only 32% say AI is reshaping their business model, 29% report that their organizations quickly adjust resources based on performance insights, and 71% believe they risk falling behind competitors.
The study also points to a group of top-performing AI leaders who see much better results, with nearly two-thirds (63%) of their AI projects delivering measurable value and successful tools rolling out across their companies in eight months rather than more than a year. According to IBM, these leaders succeed by giving AI more freedom to act on its own with 96% allowing AI to trigger actions automatically, compared to 40% of their peers, while still keeping the ability to quickly pause or adjust those systems without disrupting daily operations.
Top-performing AI leaders are also much more likely to build AI directly into daily workflows (nearly all versus 59% of peers) and enforce stricter spending discipline. Within this group, 94% base funding and staffing decisions on actual business results, 96% cut or reduce spending on underperforming projects regardless of senior sponsorship (compared to 46% of peers), and 95% reinvest productivity gains into broader business improvements (compared to 44%).