Accenture: AI Budgets Are Rising, but Only 23 Percent of Companies Report Enterprise-Wide Value

Accenture published a survey on July 26 covering 3,000 senior executives and 3,000 employees. Data was collected from April to June 2026 in twenty countries and nineteen industries, among companies with revenue above $500 million. Eighty-two percent of executives are increasing AI spending, but only 23 percent report sustained value across the enterprise. The share of companies reporting enterprise-wide value fell from 32 percent at the beginning of the year to 23 percent. This does not necessarily mean that the tools themselves have deteriorated. It more likely reflects a stricter view of what qualifies as a return. Accenture says 49 percent of companies are already testing or deploying systems that independently plan and execute tasks, yet most benefits remain confined to individual processes. Employees nevertheless report concrete operational gains. Eighty-one percent say the tools have increased their productivity, and 71 percent report greater job satisfaction. Seven out of ten executives say the productivity impact of autonomous systems has exceeded expectations. Even so, faster work does not automatically create value unless the company changes team capacity, service pricing, production volume, or the way customers are served. Views also differ on changing roles. Seventy-eight percent of executives expect major changes to job content within twelve months. Among employees, 57 percent already say their role has changed. Forty-one percent of executives feel very well prepared to manage the effects, compared with only 34 percent of employees. Both groups identify middle managers as the largest skills gap, cited in 36 percent of responses. Middle management determines whether new capabilities become a working process. Managers must divide tasks between people and systems, check outputs, resolve errors, and explain changes in performance assessment. Without preparation, a gap opens between executive decisions and actual operations. Training should therefore not be measured by attendance, but by whether managers can redesign a specific process and document the result. The survey also captured changes in how young people enter the labour market. Seventy-three percent of executives expect new entry-level AI roles within two years, and 52 percent anticipate hiring more beginners. At the same time, 61 percent of employees now ask a system for help before turning to a colleague. Accenture therefore warns that companies must redesign workplace learning, or young employees will gain speed without judgement. Macroeconomic uncertainty has not yet slowed spending. Seventy-three percent of executives expect higher inflation, 70 percent expect a slowdown in the global economy, and 49 percent plan to diversify supply chains. Even if an investment bubble burst, 52 percent would continue funding AI, although only ten percent currently see a significant bubble. Accenture concludes: “Moving from using AI to running an AI-powered enterprise requires business reinvention.” Terms explained Autonomous system: Software capable of planning and carrying out several connected tasks with limited human intervention. Enterprise-wide value: Demonstrable benefit extended across a significant part of the company, not limited to one pilot. Middle management: Managers between senior leadership and operational teams who translate strategy into daily work. Supplier diversification: Spreading purchases across multiple sources to reduce dependence on one partner or region.

Accenture: AI Budgets Are Rising, but Only 23 Percent of Companies Report Enterprise-Wide Value

The full article is available to users registered at Hard Skills