USD2.59 trillion — that’s what the world will spend on AI in 2026, according to Gartner. That’s a 47% jump year-over-year. For context, it’s roughly the GDP of France — deployed into silicon, software, and server farms in a single calendar year. It should worry any enterprise chief data officer.

And yet, most enterprises are barely in the room.

The money is currently flowing from the top of the tech food chain. Hyperscalers — your Microsofts, Amazons, Googles — are the ones writing the big checks. They’re building the roads. Enterprises are still debating whether to buy a car.

“Up to this point, AI spending has primarily been driven by technology companies and hyperscalers,” says John-David Lovelock, Gartner’s distinguished vice president analyst, in a press release. “Enterprises have yet to really flex their spending potential. That is coming, and 2026 will be the inflection year.”

Infrastructure first, everything else later

The biggest slice of that USD2.59 trillion pie — over 45% — goes to AI infrastructure. AI-optimized servers, network fabric and semiconductors. The unsexy plumbing that makes the glamorous stuff run. Gartner projects spending on AI-optimized servers alone will triple over the next five years, as cloud providers race to build capacity for the GenAI models and agentic workflows that enterprises will eventually throw at them.

Meanwhile, AI models are the fastest-growing subsegment in relative terms. Gartner forecasts 110% growth in AI model spending in 2026 alone, adding USD6 billion to this year’s tab as enterprises start stitching models into multistep workflows and tool suites. The number goes from USD15.5 billion in 2025 to USD32.6 billion this year, and USD59 billion by 2027.

AI cybersecurity spending is also worth watching. It nearly doubles, from USD25.9 billion to USD51.3 billion, in a single year. That’s a warning signal.

The productivity trap

Here’s the uncomfortable truth buried in Gartner’s numbers, and the reason this matters directly to CDOs: enterprises are playing it safe, and it’s going to cost them.

“Currently, organizations show limited appetite for using AI to drive disruptive enterprise change,” says Lovelock. “Instead, they favor tactical AI initiatives with incremental improvements in efficiency and productivity.”

Translation: most companies are using AI to shave minutes off meetings, not to reimagine their business model. That works — until a competitor that reimagined their model shows up.

The downstream problem lands squarely in the CDO’s inbox. You’re the one who has to explain to the board why the AI budget isn’t producing transformational outcomes — when the real answer is that nobody asked AI to transform anything. “CIOs face challenges in proving the value from AI investments and demonstrating tangible business outcomes,” says Lovelock. “Aligning AI initiatives with strategic business objectives is the essential step for success.”

That’s consultant-speak for: if your AI strategy doesn’t connect to revenue, risk, or competitive position, you’re doing it wrong.

Why CDOs should actually care

The data story here is the real one. AI spending on data infrastructure jumps from USD826 million in 2025 to USD3.1 billion in 2026, then USD6.5 billion in 2027. Nearly 8x in two years. Your data quality, your data governance, your data architecture: these are no longer back-office concerns. They are the difference between AI that works and AI that hallucinates your next board presentation.

The hyperscalers built the engine. 2026 is the year enterprises get handed the keys. But are enterprise CDOs ready?

Image credit: iStockphoto/Lacheev

Stay ahead with CDO Nexus

Join an exclusive community of CDOs and data leaders

When you join, you will:

Access curated trends and thought leadership from IBM and industry experts.Participate in private roundtables and webinars to solve regional CDO challenges.Connect with a network of like-minded leaders navigating the same data landscape.