AI spending is outrunning organizational readiness.
Reports dated Jan. 23, 2025, Sept. 16, 2026, and 15 Sep 2026 describe a familiar corporate reflex: invest first, work out the operating model later. Seventy-nine percent of executives say their organizations increased AI investment over the past ten months, with spending up by an average of 31%.
That surge is moving faster than many organizations can absorb it. Only 24% say they provide comprehensive AI enablement, while 27% say employees are largely left to learn and adopt AI on their own. Apparently, handing people new systems and hoping for workplace alchemy remains a management strategy.
The technology is not the main bottleneck
Seventy-seven percent of executives agree that AI value is constrained more by people and processes than by technology. That finding cuts through the usual hardware-and-model obsession: organizations are not only buying AI tools, they are trying to fit them into decision-making, workflows, training, and accountability systems that were not designed for them.
Srdjan Jovanovic, Chief People Officer at HTEC, puts the problem in operational terms: “Organizations often focus on teaching people how to use AI. The bigger challenge is redesigning the environment in which they use it.” Training matters, but it cannot fix a process that gives employees no clear rules, useful feedback, or way to measure results.
The cost of that gap is visible in the numbers. Twenty-four percent of organizations experienced higher AI usage costs because they lacked sufficient tracking, while between 18% and 21% reported reduced productivity, lower employee confidence, or slower innovation after adopting AI.
Those outcomes do not prove that AI fails. They show that investment alone does not create value. Marko Anić, VP, Engineering and Delivery at HTEC, says many organizations are investing heavily in AI but still lack “a reliable view of the value it delivers across different functions, teams, and use cases.” Without that view, companies cannot tell which tools deserve more funding, which processes need redesign, or where usage has become expensive noise.
Executives want impact, not just lower costs
Cost reduction ranks among the most important AI outcomes for only 18% of executives. Their priorities point elsewhere: decision quality leads at 29%, followed by productivity gains and better customer experience at 27% each, innovation at 26%, and faster time-to-market at 25%.
That list explains why tracking matters. An organization cannot judge decision quality, customer experience, or innovation with a single software bill. Each outcome needs a clear measure, a responsible team, and a process that connects AI use to business results. Without those links, ambitious targets become corporate wallpaper.
Alex Rumble, Chief Marketing Officer at HTEC, describes the shift this way: “Just over a year ago, much of the AI conversation was about experimentation. Today, organizations are scaling their initiatives, investing more, and setting more ambitious expectations for the value AI should create. But higher investment does not automatically lead to greater impact.” The figures support the warning: adoption is scaling before enablement and measurement have caught up.
The safety debate adds another layer to the corporate rush. A rare consensus among AI industry leaders—including Dario Amodei, CEO and co-founder of Anthropic, Sam Altman of OpenAI, and Elon Musk of SpaceXAI—calls for slowing AI development because of safety concerns. Organizations are therefore being asked to move quickly on investment while leaders argue that development itself needs restraint. That is not a contradiction the balance sheet can solve.
HTEC’s report captures the central tension: “Even at this early stage, one pattern already stands out: organizations are committing capital to AI faster than they are building the capabilities needed to measure and govern outcomes effectively.” The next phase of AI adoption will not be judged by how many tools companies purchase. It will be judged by whether employees know how to use them, managers can track their effects, and leaders can stop paying for systems that create more friction than value.
Based on
