Only seven percent of organizations are truly ready for AI. This is according to Veeam’s new Data and AI Trust Gap report, presented at VeeamON London. Of the 600 senior executives surveyed, 88 percent are already using or testing AI agents, while 95 percent indicate that data issues have already slowed AI progress.
The survey, conducted among executives in financial services, healthcare, manufacturing, retail, and technology, makes it clear that AI adoption is proceeding significantly faster than the governance structures designed to manage it. Only 28 percent of organizations are confident they can detect AI systems operating outside approved parameters.
This has implications for how organizations handle AI errors. Of the organizations currently using AI, only 29 percent can identify within minutes which systems an AI agent has accessed. For 22 percent, it is quickly traceable which data was used. Only 40 percent are confident they can isolate an agentic AI error and apply precision remediation.
Boardroom versus operational reality
There is a striking gap between what CEOs think and what IT leaders experience. 65 percent of CEOs believe they have a complete overview of AI, compared to just 48 percent of technical leaders. Opinions also differ on data accountability. 52 percent of CEOs claim to actively drive decisions based on data, but only 41 percent of CISOs and 38 percent of CIOs share that view.
Veeam CEO Anand Eswaran sums it up as follows: “Most organizations don’t have an AI adoption problem; they have an AI trust problem.” The next phase of AI, he argues, will be driven by trust, not by infrastructure investment or experimentation.
On the other hand, AI-ready organizations achieve significantly better results. Among fully AI-ready organizations, 97 percent report measurable business benefits from data and AI investments. On average, that figure is just 48 percent.
Shadow AI and External Regulatory Pressure
Within organizations, unauthorized AI use has now become mainstream. 95 percent report shadow AI, and 93 percent consider this a significant risk. Yet only 25 percent offer an approved alternative. 44 percent cite increased cyber risk as the greatest danger of shadow AI.
Regulatory pressure is growing from the outside. 61 percent indicate that the EU AI Act has already influenced investment strategies over the past twelve months. 47 percent cite maintaining audit trails for AI decisions as their biggest compliance concern.
Clear ownership appears to be decisive. In organizations where the CISO has ultimate responsibility for AI agent risk, the likelihood of detecting anomalous AI behavior is 24 percent higher. With shared ownership, that likelihood is 47 percent lower.