AI adoption reaches 88%, while readiness stands at 7%
Artificial intelligence adoption is advancing rapidly across organizations, but the ability to govern, oversee and secure these systems is not evolving at the same pace.
A new study from Veeam Software reveals that while 88% of organizations already use or test AI systems, only 7% can be considered truly prepared to govern them effectively.
Artificial intelligence is being integrated into business processes at a rapid pace, but organizations’ readiness to manage it securely appears to be progressing much more slowly.
That is the conclusion of the Data and AI Trust Gap report published by Veeam Software, which identifies a growing gap between the adoption of AI tools and organizations’ ability to oversee, govern and control the risks associated with these technologies.
According to the study, 88% of organizations already use or are testing AI agents, while only 7% can be considered truly prepared to operate AI systems in a secure and controlled manner.
The research was based on a survey of 600 senior executives across industries including financial services, healthcare, manufacturing, retail and technology.
The challenge is no longer AI adoption.
One of the report’s central findings is that most organizations have moved beyond the experimentation phase and are actively implementing AI initiatives.
However, operational readiness, data governance and control mechanisms are not advancing at the same pace.
In fact, 95% of surveyed organizations said data-related issues are already slowing down their artificial intelligence initiatives.
According to Anand Eswaran, CEO of Veeam Software:
”Most organizations do not have an AI adoption problem; they have an AI trust problem.”
According to Eswaran, the challenge for businesses is no longer simply deploying artificial intelligence, but ensuring that the data used by these systems is secure, verifiable, controlled and recoverable in the event of an incident.
A gap between perception and reality.
The report also identifies significant differences between executive leadership and the teams responsible for implementing technology initiatives.
While 65% of chief executives believe their organizations have fully developed AI capabilities, only 48% of technical leaders share that assessment.
Similarly, 52% of CEOs believe they actively lead data management efforts, a perception that is not fully shared by technology and information security leaders.
When AI fails.
The research warns that risks associated with artificial intelligence may not manifest as traditional system outages.
As AI agents gain greater autonomy, failures can originate from the data they use, the decisions they make or the actions they execute.
According to the study:
Only 29% of organizations can quickly identify which systems an AI accessed.
Just 25% can determine what actions it performed.
Only 24% can establish which decisions it influenced.
Just 22% can accurately identify which data it used.
In addition, only 40% of business leaders said they are highly confident in their ability to isolate and precisely reverse a failure caused by autonomous AI systems.
The rise of shadow AI.
Another trend identified in the report is the growing use of unauthorized artificial intelligence tools within organizations.
Ninety-five percent of organizations acknowledge some level of informal or unauthorized AI use by employees, while 93% consider the situation a significant risk.
However, only 25% provide approved corporate alternatives for using these tools.
Trust as a competitive advantage.
The study concludes that organizations that establish clear governance, control and data resilience mechanisms achieve stronger results from their artificial intelligence initiatives.
Among organizations classified as fully prepared for AI, 97% reported measurable business benefits from their investments in data and artificial intelligence, compared with 48% across the broader sample.
According to Veeam, the next phase of artificial intelligence adoption will be less about technological experimentation and more about organizations’ ability to build trust, traceability and control over the data that powers these systems.
|
|
