Islands of productivity hide in a sea of noise. In 2026, the winners will transform fragmented AI experiments into governed, data-driven decisions.
AI’s Hidden Value Problem
AI is everywhere—but its impact is invisible. Mike Capone argues that the debate about whether AI “delivers value” misses the point. “Boardrooms keep hearing that only a tiny fraction of enterprises are truly ‘AI-ready,’” he says. “But walk the halls of any large company and you’ll find people quietly using AI in documents, slides and code every day. That’s real productivity. Right now, it’s dark matter—it shapes outcomes, but it’s invisible to the P\&L, rarely based on complete data, and entirely outside governance.”
Research backs this up: Boston Consulting Group reports only 5% of enterprises are structurally prepared for an AI future. Qlik’s own analysis shows the same paradox—AI is everywhere, yet its value rarely survives the jump into audited metrics or risk models.
“The real problem isn’t whether AI delivers value—it’s that most enterprises are underachieving.”
— Mike Capone, CEO, Qlik
The 2026 Imperative: Controlled Decentralization
Capone predicts that 2026 will be the year enterprises close this gap. Instead of betting on a single model or platform, organizations must design architectures that allow them to swap tools without rewriting business logic or losing control of data. “The old pendulum between tight central control and chaotic self-service is breaking down,” he says. “The companies we see winning are practicing controlled decentralization: they keep definitions, governance and sovereignty non-negotiable, and push experimentation and automation out to the teams closest to the work.”
Intelligence as a Utility
Qlik expects intelligence to behave more like a utility than a feature. Smaller models, local inference, and edge computing will push decisions closer to where data lives—in factories, stores, vehicles, and devices—while shared analytics remain the anchor for what “good” looks like. “Over time, the price of intelligence per decision will go down, and the expectations for accountability will go up,” Capone notes. “The winners will be the companies that can turn hidden, ad-hoc AI usage into an explicit, governed system of decisions, where every agent, assistant and application stands on the same trusted data and analytics.”
