Chief AI officers lack control over AI spending: study
What high-impact AI leaders do differently
Some organizations have found a way through. IBM identified a group of high-impact CAIOs whose organizations report stronger results: 63% of their AI projects deliver measurable business outcomes, versus 43% elsewhere, and they scale initiatives enterprise-wide in eight months rather than 13.5. The data is self-reported and shows association, not cause.
The clearest difference is financial discipline. Some 96% of these leaders cut or stop funding for underperforming AI initiatives, even with senior sponsorship, compared with 46% of others.
“We stop when the value isn’t there,” said Pascal Belaud, executive vice-president and chief AI and data officer at Truist.
What should a chief AI officer be?
That discipline points to the kind of leader the report describes. Wright wrote that successful AI leadership “requires more than technical expertise or operational oversight,” adding: “It requires gathering evidence, influencing decisions, and measuring success.”
IBM’s action guides show what that means in practice. CAIOs should define which decisions AI can make and which need human judgment, redesign high-value workflows end-to-end around AI, and manage AI as a portfolio, funding what works and stopping what doesn’t.