A survey by Certinia, a cloud service management consultancy, indicates that only one-fifth of companies that adopted artificial intelligence say the technology exceeded expectations. The data, published on August 22, points to a mismatch between the promise sold to executives and the actual delivery in operations.
The study heard from leaders and technology professionals in organizations of varying sizes. While top management often publicizes AI success stories, the teams that implement and maintain the systems report results below expectations — which erodes internal trust.
This gap between leaders' perception and operators' practice is familiar from previous hype cycles in technology, but now takes on critical contours because investments are on a large scale. Certinia points out that investors are beginning to question when the return will materialize, and the pressure for immediate results could lead to cuts in projects that have not yet matured.
For those running systems in production, the message is direct: AI success metrics must be agreed upon before deployment, not after. Without objective indicators, the risk of disappointment among leadership and abrupt project discontinuation is high.
The newspaper understands that the problem is not the technology itself, but the unrealistic expectations surrounding it. AI projects in production need measurable goals from the start, not transformation narratives that fail to hold up on Monday-morning delivery.