In the collective consciousness, artificial intelligence (AI) is kind of modern-day sorcery. From writing persuasive business proposals to uncovering blindspots in finance reports, distilling down years of historical data into pure-gold analytics to seeing into the future – AI can make all magical things happen in a heartbeat.
So, it’s little wonder that companies are going out on a tear to adopt it to elevate their business chops. But a newly released market research speaks a slightly different story.
A Futurum survey finds that while 75% CEOs believe that AI will drive up their competitive advantage, only a meagre 25% feel ready to adopt it.
The study, a joint project of Futurum and Kearney, peers into the minds of CEOs to understand how close AI is on their roadmap, what bumps they see ahead of them, and measure their real AI readiness.
Conducted between November 2024 and January 2025, the survey polls 213 CEOs in top companies across North America, Europe and Asia with turnovers of over billion dollars, to understand the ground truths of the AI buzz.
Dion Hinchcliffe, VP of CIO practice and one of the authors and key figures behind the survey, notes, “There’s a big disconnect between the C-suite leaders and a lot of technology staff, and we can actually use this to help deconflict.”
According to the findings, 59% CEOs believe that they are leading AI in their organization in strategy, but not in implementation. Interestingly, in high-performing firms, CEOs self-reported that they are taking a decentralized approach to AI where larger goals and visions are set for departments to pursue while the CEOs themselves maintain oversight from a distance.
In less performing firms, however, CEOs are seen to be more hands-on, and their micromanaging the campaigns is leading to significantly higher failure rates.
It also finds that half (50%) of the CEOs favor a fast-follower approach where they observe and learn before going into scaling. By contrast, companies that aggressively pursue AI, 58% said they are struggling.
“I’ve talked to over 250 CIOs since generative AI came out and almost universally, they say the board is breathing down their neck to do something before a competitor derails what they’re doing with AI. That’s the big fear,” Hinchcliffe says.
They have a sense of what the short- and the long-term roadmaps should be, he found, but the things they want to do with AI is not ready.
“Companies feel the pressure to adopt and implement AI, but the path to ROI is not yet obvious,” says Daniel Newman, CEO of The Futurum Group. “Companies are investing in PoCs and turning to their software and IT providers for help, but just because the technology is there doesn’t mean the right use cases are evident.”
Driven by FOMO but squeezed by scant budgets, many of the companies are taking a shortcut to success. Hinchcliffe found that only 19% of the CEOs aim to use AI for true transformation compared to the 78% that claims confidence in the technology.
“I think it’s really not time to do major AI transformation,” he opines, “because the vendors are not mature or ready.”
The study finds that 95% of the companies focus on quick wins vs building breakthrough products in AI, and concludes that this short-term thinking may lead to missed opportunities for the companies.
While presenting the report at the AI Field Day event in January, Hinchcliffe shared some real-world AI transformation stories to drive the point home. When used for full-scale transformation, AI has proven to unlock breakthrough progress in many cases. According to the research, a banking organization was able to make significant cuts (60%) to fraud cases using AI risk models. In another case, a manufacturer reduced downtime (40%) with AI-led predictive maintenance.
These success stories go to show that if technology and resource are allocated right, significant competitive advantage can be gained.
Hinchcliffe also says that the true value of AI can only be unlocked by balancing immediate impact with long-term breakthroughs. However, he cautions that “AI without ROI tracking is not worth doing.”
Significant blind spots exist within company strategies that are leading to unrealized ROI in AI initiatives. The study finds that 48% of high achievers track ROI and only 17% of those that do report to be struggling to achieve the desired ROI.
“Well-managed organizations that have more rigor tends to be larger organizations that have a professional project management office and IT departments that actually track results,” Hinchcliffe said.
Another significant stumbling block, the survey finds, is talent shortfall. 57% CEOs said that there is not sufficient AI expertise within firms. In effect, high-performing companies are now demanding AI literacy as a default requirement across all roles.
Other findings include low data-readiness which is linked to 60% of the failed effort in AI, and a deep cultural divide that is compelling CEOs to get directly involved in the change management.
“The culture of rapid technological change is always significant,” says Newman. “AI is driving massive organizational change that impacts how people work and the tools they use. This type of change always creates friction in cultures making it even harder for leaders to execute.”
A worrying lag is also found to exist in AI governance and ethics. 80% CEOs said that AI bias and ethics are a concern for them, yet half of them still lacks a governance framework within the firm. The industries where AI guardrails are noticeably enforced are finance and healthcare.
AI may have percolated through a majority of business houses, but its highest concentration still continues to be in the biggest corporations, and generally at the corners of small enterprises. In 2025, CEOs are ostensibly focusing their efforts on pushing the technology broadly into the organization but they need a vision that transcends the myopic ambitions that keep AI projects within the circle of doom, and pushes them out into daylight.
“[Companies] need to infuse AI in their organization, find the right place to do it without blowing everything up because AI has real risks,” Hinchcliffe said.
