IBM's New CEO Study: Nearly Half of Operational Decisions Will Be Made by AI by 2030

Right now, 25% of codifiable operational decisions in the enterprise are handled by AI. By 2030, that share will nearly double to 48%, with zero human involvement. That's the headline from IBM's annual CEO study, released today.

IBM CEO Study on AI decisions: the key numbers

The IBM Institute for Business Value, working with Oxford Economics, surveyed 2,000 CEOs across 33 countries and 21 industries. The takeaway is blunt: AI is no longer a side project. CEOs are restructuring their entire leadership teams around it.

Among the standout data points: 64% of CEOs now feel comfortable basing major strategic calls on AI-generated input. Meanwhile, 83% say AI sovereignty is essential to their business strategy.

The Chief AI Officer goes from rare to standard

Here's the stat that should grab your attention: 76% of surveyed organizations now have a Chief AI Officer, up from just 26% in 2025. That's a threefold jump in a single year. Still, some analysts compare this surge to the Chief Digital Officer wave a decade ago, a role that often faded once digital became everyone's job.

Companies with an AI-first C-suite structure scaled 10% more AI initiatives than peers. Correlation or causation? Probably a bit of both, but the signal to boards is clear.

Massive reskilling ahead: 53% of workers need new skills

Between now and 2028, the study projects 29% of employees will need to shift roles entirely, and 53% will require upskilling just to keep their current job. Yet 83% of CEOs believe success with AI depends more on people's adoption than on the technology itself.

The gap is stark: only 25% of workers actually use AI regularly, despite 86% of CEOs believing their teams already have the skills. That disconnect between boardroom optimism and shop-floor reality is one of the biggest barriers to scaling AI, and this study makes it painfully obvious.

What IBM's AI decisions study means for you

If you work in any sizable company, the writing is on the wall. The question has shifted from "should we use AI?" to "who governs it?" Google already has 75% of new code written by AI. The execution phase is here.

The open question IBM doesn't fully answer: is handing 48% of operational decisions to algorithms, even guardrailed ones, truly the right call? Governance frameworks will separate the winners from the cautionary tales.