Seventy-seven percent of business decision-makers believe their companies are putting too much money into AI projects that only make small tweaks to existing processes, according to a survey commissioned by Avanade, a Microsoft-focused IT advisory services provider. Just 11% of these leaders expect artificial intelligence to fundamentally reshape how their organizations function, generate value, or compete over the next year. The findings suggest widespread frustration with IT departments that haven't moved past using AI for minor adjustments to current workflows.
Nearly two-thirds of respondents say AI hasn't yet produced returns that justify the money and effort invested, and three-quarters believe their organizations have overstated how mature their AI programs actually are. The survey included C-suite executives, directors, and managers responsible for business decisions. Only 11% anticipate AI will fundamentally alter their organization's operations, value creation, or competitive position in the coming year.
Ben Beath, Avanade's chief AI officer, says many organizations and their CIOs haven't yet embraced AI's full potential despite recognizing what's possible. "Confidence is high, but confidence isn't the same as change," he notes. The report argues that when business leaders don't see major returns from AI spending, it indicates organizations haven't reached AI maturity, likely because of misguided strategy and approach. Beath explains that adding AI to broken or outdated processes simply creates "a faster legacy process."
The report identifies a leadership gap as the core problem: most organizations suffer from a crisis where CIOs pursue AI projects without broader executive involvement. True transformation requires fully integrating AI into business operations rather than attaching it to existing workflows, according to Beath. He advises CIOs working with other business leaders to ask what company they would build from scratch with AI today, rather than simply identifying where to add AI to current operations. The report notes that AI transformation needs someone thinking across data, applications, technology, processes, and business strategy together, not just the technical pieces. Several organizations featured in the report have found success combining both incremental and transformational AI uses, treating small improvements as stepping stones while simultaneously pursuing projects that fundamentally change how work gets done, though challenges around data quality, legacy systems without APIs, and risk management continue to hold back wider transformation efforts.
The report concludes that IT leaders focused on truly transformational AI uses rather than isolated efficiency gains can shift the tide at their organizations, bringing greater maturity and better business results. Organizations should reimagine business processes with AI embedded at every step instead of asking how to layer AI incrementally on top of existing workflows. The bottom line: having Copilot licenses and a few dozen use cases might feel like progress, but it probably isn't maturity if it hasn't changed how work actually gets done. The tension revealed here points to a broader reckoning about whether technology investments serve the org chart or the customer, and whether executives are willing to dismantle working processes to discover what AI can actually unlock. Four years into this era, impatience is no longer just a budget problem—it's becoming a competitive one.