Most Companies Have AI Ownership Backwards: Why AI Doesn't Belong in IT
Despite tens of billions in collective investment, a staggering 95% of enterprise AI pilots deliver zero measurable P&L impact, according to MIT’s Project NANDA.[1] The instinct is to blame the technology, the vendors, or the talent shortage, but the data points to a different culprit: organizational design.
Most companies handed ownership of AI to IT, but AI requires businesses to transform the way they work. And businesses aren’t transformed by IT.
To unlock value, organizations need to rethink ownership:
- Shift accountability to business units to define use cases, success metrics, and outcomes
- Adopt a portfolio mindset to scale high-impact initiatives and stop underperforming ones faster
- Align governance with transformation goals, not traditional IT operating models
If most AI pilots are failing, the answer isn’t better pilots. It’s getting ownership right. The organizations that win will be those that connect AI investment directly to business performance.
Read the full article to learn how to redesign AI ownership and turn pilots into measurable impact.
[1]MIT NANDA, The GenAI Divide: State of AI in Business 2025 (MIT 2025).
