itstargetnews
The chief data officer (CDO) has evolved from a niche compliance role to one of the most critical positions for AI deployment. These executives now sit at the intersection of data management, AI strategy, and workforce readiness. Their decisions determine whether businesses move from AI pilots to scale production or remain stuck in experimental mode.
So the third annual Informatica survey – the largest survey of CDOs specifically on AI readinesswhich spans 600 executives worldwide – has particular weight. The findings reveal a dangerous disconnect that explains why many organizations struggle to scale AI beyond pilots: While 69% of businesses deploy generative AI and 47% run agent AI systems, 76% admit that their management frameworks cannot keep up with how employees use these technologies.
The survey revealed what Informatica calls a "trust paradox" – and explains why data leaders are dangerously overconfident about AI readiness. Organizations are deploying generative AI systems faster than they are building the management and training infrastructure to support them. The result: Employees often rely on the data that powers AI systems, but organizations recognize that their workforce lacks the literacy to question data or use AI responsibly. Seventy-five percent of data leaders say employees need improvement in data literacy. Seventy-four percent require AI literacy training for day-to-day operations.
"The gap now is, can you trust the data to set an agent loose on it?" Graeme Thompson, CIO of Informatica, told VentureBeat. "Agents do what they are supposed to do if you give them the right information. There is such a lack of confidence in the data that I think that is the gap."
Why infrastructure is not the bottleneck for data and AI
GenAI adoption jumped from 48% last year to 69% today. Nearly half of organizations (47%) now run agent AI – systems that operate autonomously rather than simply generating content. This rapid expansion has created a race to acquire vector databases, upgrade data pipelines, and expand computational infrastructure.
But Thompson dismisses infrastructure gaps as the main problem. The technology is there and working. The limitation is organizational, not technical.
"The technology that we have now, the infrastructure, more than — this is not the problem," Thompson said. He compared the situation to amateur athletes who regret their equipment. "There is a long way to go before the equipment is the problem in the room. People chase equipment like players. Those golfers are a sucker for a new driver, a new putter that will fix their physical inability to hit the golf ball straight."
The survey data supports this. When asked about investment priorities in 2026, the top three are all people and process issues: data privacy and security (43%), AI management (41%), and workforce development (39%).
Five hard lessons for enterprise CDOs
The survey data combined with Thompson’s implementation experience revealed specific lessons for data leaders trying to transition from pilots to production.
Stop chasing infrastructure, fix people’s problems
The trust paradox exists because organizations can deploy AI technology faster than they can train people to use it responsibly. Seventy-five percent need improvement in data literacy. Seventy-four percent need AI literacy training. A technology gap is a people gap.
"It’s easier to get your people who know your company and know your data and know your processes to learn AI than to bring in an AI person who knows nothing about things and teach them about your company," Thompson said. "And also AI people are more expensive, just like data scientists are more expensive."
Make the CDO an execution function, not an ivory tower
Thompson built Informatica so that the CDO reported directly to him as the CIO. This makes data management an implementation function rather than a separate strategic layer.
"That was a deliberate decision based on that function of getting things done instead of an ivory tower function," Thompson said. The structure ensures that data teams and application owners have the same priorities through a common boss. "If they have a common boss, their priorities should be aligned. And if not, it’s because the boss isn’t doing his job, not because the two functions aren’t working on the same priority list."
If 76% of organizations cannot manage the use of AI effectively, the reporting structure may be part of the problem. Siled data and IT functions create conditions for pilots that cannot be measured.
Build literacy outside of IT teams
The breakthrough insight is that AI literacy programs must extend beyond technology teams to business functions. At Informatica, the chief marketing officer is one of Thompson’s most powerful AI partners.
"You need that literacy in your business teams as well as your technology teams," Thompson said.
He noted that the marketing operations team understands technology and data. It knows the answer to "How can I get more value from my limited annual marketing program dollars?" is by automating and adding AI to how that work is done, not adding people and more Google ad dollars.
Business-side literacy creates a pull rather than a push for AI adoption. Marketing, sales and operations teams are beginning to demand AI capabilities because they see strategic value, not just efficiency gains.
Treat AI as strategic expansion, not cost reduction
Data leaders have spent decades battling perceptions that IT is just a cost center. AI offers the opportunity to change that narrative, but only if CDOs reframe the value proposition from productivity savings.
"I’m very disappointed that, given this new technology capability on a plate, as IT people and as data people, we immediately turn around and talk about saving productivity," Thompson said. "What a waste of time."
The tactical shift: Pitch AI’s ability to remove headcount constraints entirely instead of reducing existing headcount. It reframes AI from operational efficiency to strategic capability. Organizations can expand market reach, enter new geographies and test initiatives that were previously cost prohibitive.
"It’s not about saving money," Thompson said. "And if that’s the main approach you have, then your company won’t win."
First vertically, scale the pattern
Don’t wait for fully horizontal data management layers before delivering production value. Choose a high-value use case. Build a complete governance, data quality and literacy stack for that specific workflow. Validate the results. Then copy the pattern to the adjacent use cases.
It adds value to production as the organization’s capability improves.
“I think this space is moving so fast that if you try and solve 100% of your management problem before you get your semantic layer problem, before you get your glossary of terms problem, then you won’t create any results and people will lose patience," Thompson said.
