The Chief Data Officer (CDO) has evolved from a niche compliance role to one of the most critical positions for AI deployment. These leaders now stand at the intersection of data governance, AI strategy, and workforce readiness. Their decisions determine whether companies move from AI pilots to production scale or remain stuck in experimentation mode.
That’s why Informatica’s third annual survey… largest survey to date of CDOs specifically on AI readinessbringing together 600 leaders around the world, has particular weight. The findings reveal a dangerous disconnect that explains why so many organizations are struggling to scale AI beyond pilot projects: while 69% of companies have deployed generative AI and 47% are using agentic AI systems, 76% admit that their governance frameworks can’t keep pace with how employees actually use these technologies.
The investigation reveals what Informatica calls a "trust paradox" – and explains why data leaders are dangerously overconfident about AI readiness. Organizations have deployed generative AI systems faster than they have built the governance and training infrastructure to support them. The result: Employees generally trust the data that powers AI systems, but organizations recognize that their people lack the knowledge to question that data or use AI responsibly. Seventy-five percent of data leaders say employees need to develop their data literacy skills. Seventy-four percent need AI training for their daily operations.
"The gap now is: can you trust the data to release an agent?" Graeme Thompson, CIO at Informatica, told VentureBeat. "Agents do what they’re 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 reason for the gap."
Why infrastructure is not a bottleneck for data and AI
GenAI adoption has increased from 48% a year ago to 69% today. Nearly half of organizations (47%) now use agentic AI, which is systems that take action autonomously rather than simply generating content. This rapid expansion has created a race to acquire vector databases, upgrade data pipelines, and expand compute infrastructure.
But Thompson sees infrastructure gaps as the biggest problem. The technology exists and works. The limitation is organizational and not technical.
"The technology we have now, the infrastructure, is more than — that’s not the problem yet," » said Thompson. He compared the situation to amateur athletes blaming their equipment. "There’s a long way to go before equipment becomes a problem in the room. People chase equipment like golfers. These golfers love a new driver, a new putter that will cure their physical inability to hit a golf ball straight."
The survey data confirms this. When asked about investment priorities for 2026, the top three questions relate to people and processes: data privacy and security (43%), AI governance (41%), and upskilling the workforce (39%).
Five Hard Lessons for Corporate CDOs
The survey data, combined with Thompson’s implementation experience, reveals specific lessons for data managers trying to move from pilot to production.
Stop chasing infrastructure, solve the people problem
The trust paradox exists because organizations can deploy AI technology faster than they can train their employees to use it responsibly. Seventy-five percent need to improve their data literacy skills. Seventy-four percent need AI training. The technological gap is a human gap.
"It’s much easier to get your people who know your business, your data, and your processes to learn AI than to bring in an AI person who knows nothing about these things and teach them about your business." » said Thompson. "And AI specialists are very expensive, as are data scientists."
Make the CDO an execution function and not an ivory tower
Thompson structures Informatica so that the CDO reports directly to him as CIO. This makes data governance an execution function rather than a separate strategic layer.
"This is a deliberate decision based on the fact that this function is a getting things done function rather than an ivory tower function," » said Thompson. The structure ensures that data teams and application owners share common 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 list of priorities."
If 76% of organizations fail to effectively manage the use of AI, the hierarchical structure could be part of the problem. Siled data and IT functions create the conditions for pilot projects that never evolve.
Develop literacy outside of IT teams
The revolutionary idea is that AI induction programs must extend beyond technology teams to encompass business functions. At Informatica, the CMO is one of Thompson’s strongest AI partners.
"You need this knowledge within your sales teams as well as your technology teams," » said Thompson.
He noted that the marketing operations team understands technology and data. He knows that the answer to "How can I get the most out of my limited marketing program budget each year?" it’s by automating and adding AI to how this work is done, without adding people or more Google ad dollars.
Literacy on the business side creates a pull rather than a push for AI adoption. Marketing, sales, and operations teams are starting to demand AI capabilities because they see strategic value, not just efficiencies.
Present AI as a strategic expansion, not a cost reduction
Data industry leaders have spent decades battling the idea that IT is just a cost center. AI offers the opportunity to change this narrative, but only if CDOs reframe the value proposition away from productivity gains.
"I’m very disappointed that given this new technological capability, as computer scientists and data scientists, we immediately turn around and talk about productivity savings." » said Thompson. "What a wasted opportunity."
The tactical shift: Pitch AI’s ability to remove headcount constraints entirely rather than reducing existing headcount. This reframes AI from operational effectiveness to strategic capability. Organizations can expand their market reach, enter new geographies, and test previously cost-prohibitive initiatives.
"It’s not about saving money," » said Thompson. "And if that’s primarily your approach, then your business won’t win."
First go vertical, resize the pattern
Don’t wait for perfect horizontal layers of data governance before generating production value. Choose a high-value use case. Create the full governance, data quality, and literacy stack for this specific workflow. Validate the results. Then replicate the model in adjacent use cases.
This generates production value while gradually building organizational capabilities.
“I think this space is moving so quickly that if you try to 100% solve your governance problem before you get to your semantic layer problem, before you get to your glossary of terms problem, then you’ll never get any results and people will lose patience." » said Thompson.
