Companies have more than doubled their spending on artificial intelligence in less than a year as AI moves from an experimental technology project to a broader business investment, according to new research by Boston Consulting Group (BCG).
Corporate AI spending has risen from about 1.7 percent of revenue in late 2025 to 3.3 percent, with more than 80 percent of the spending now taking place outside enterprise IT budgets, BCG said in its Applied AI Index 2026.
The research, based on a survey of 1,330 C-suite executives and senior leaders, also found that nearly half of companies are now generating meaningful value from AI, challenging the view that corporate AI investment has yet to translate into returns.
Michael Grebe, managing director and senior partner at BCG, said companies that continue to treat AI solely as an IT expense are underestimating both the amount being spent and the potential returns.
The shift in spending is also changing where companies are building AI capabilities.
Jeff Walters, managing director and senior partner at BCG, said non-technology companies were increasingly developing capabilities that had previously been concentrated in technology businesses.
He cited product management and workflow design as examples, with companies needing people who can determine which processes should change and what AI agents should be built to perform them.
“The agentic and software toolkit is now so important, no matter what industry you’re in,” Walters said during a briefing on the report.
Read also: Technology, economic realities are reshaping modern consumer, Sanwo-Olu reminds brand owners at CVA
He said the change was evident across industries, including mining and offshore oil, where AI capabilities are becoming relevant even though the companies were not traditionally technology-driven.
The growing investment is also expected to alter corporate workforces. Companies surveyed by BCG expect overall headcount to decline by about 10–15 percent by 2030, while the share of workers in dedicated AI-related roles is expected to rise from seven percent in 2026 to 22 percent.
Walters said the figures should not be interpreted as a forecast that economies will lose 10–15 percent of jobs.
Rather, they reflect what companies currently expect based on their experience with AI, with the eventual impact dependent on how work and new roles evolve.
The research suggests that the pressure will be concentrated more heavily on coordination and middle-management tasks, while demand grows for workers who can design AI workflows, build AI agents and manage the architecture and controls around them.
The rapid deployment of AI is creating another challenge: governance.
BCG found that 42 percent of companies expect to give AI agents genuine decision-making authority by 2030, while only five percent currently have the full set of controls needed to deploy such systems safely.
Walters said companies would have to strengthen controls around autonomous AI without allowing governance requirements to become a bottleneck for deployment.
The findings indicate that AI is increasingly being treated as an enterprise-wide investment rather than a technology function, with companies changing how they allocate capital, organise work and build capabilities around the technology.
Ayomide Odunlami is a Tax Reporter at BusinessDay, covering Nigeria’s tax reforms, compliance trends, and government revenue strategies. She reports on how evolving tax policies affect businesses, investors, and the broader economy, providing clarity on complex regulatory issues through data-driven journalism.
