At this year’s Gartner IT Symposium, it was apparent that the AI buzz hasn’t died down. However, it was also clear that CIOs have changed how they view the technology.
The era of fascination is over.
What we saw instead were rooms full of leaders who are pragmatic and purposeful, while maybe becoming a tad impatient. The shift is on and tech executives want to know how to scale AI, govern it responsibly, and extract measurable value.
And while the sessions delivered nothing but insight, some of the most revealing perspectives came from the hallways, quick conversations, and casual exchanges with the people leading AI adoption at scale.
These are frontline reflections from CIOs and those working directly alongside them.
Building the Right Foundations
Before AI can deliver, CIOs are realizing they need to clean house.
“You have to get your house in order,” said Erik Bakstad, CEO of Ardoq. “Understand the data you have, the opportunity you have, and the context it lives in.” This back-to-basics sentiment echoed through the show floor, a quiet pushback to years of unchecked experimentation.
Clay Ritchey, CEO of Verato, added: “Before you can think about all the wonderful things AI can do, you have to focus on foundational investments, so you have high-quality, high-fidelity data.”
“Understand the data you have, the opportunity you have, and the context it lives in.“
Erik Bakstad, Ardoq
Identity resolution, in particular, is becoming a non-negotiable step in preparing systems for AI integration. That preparation also requires strong architectural discipline.
Inspi founders, Klaus Isenbecker and Kadir Özbayram, warned that CIOs risk losing control if they let AI develop without governance. “If you don’t govern it, your data will be everywhere,” said Özbayram. Without that visibility, they suggested, organizations aren’t leading with AI, they’re just hoping it doesn’t go off track.
Vultr CMO, Kevin Cochrane pointed to resiliency and redundancy as top priorities.
“We cannot be dependent on one single point of failure,” he said, referencing a real-world incident where a single engineer’s misconfigured database table disrupted global operations. His advice was to think multi-cloud and make sure your GPU architecture is as distributed as your applications.
Security leaders are also sounding the alarm.
Wellington Estevao, Senior Solutions Engineer at Saviynt, emphasized an often overlooked reality: in the rush to innovate with AI, many organizations still lack visibility into who is accessing what, and why. “You can’t govern what you can’t see,” he said. “And if you don’t govern it, identity becomes your biggest risk.”
AI at Work
The tone has shifted from “what if” to “how fast.”
“Don’t over-rotate on replacing people with AI agents,” cautioned Tal Klein, CMO at Lakeside. “Think more about improving the outcomes you’re delivering.” The value of AI, he emphasized, lies in enhancing productivity, even if the public narrative is still focused on workforce reduction.
That message resonated across multiple sectors.
As leaders at Anunta shared, they believe self-healing endpoints are going to be the future. By designing systems that automatically resolve issues without the user even knowing, they see a path toward more proactive, resilient operations. And while automation plays a key role, the intent isn’t to replace people. It’s to let them focus on more valuable work.
“If you don’t govern it, your data will be everywhere.”
Kadir Özbayram, Inspi
Micha Kiener, CTO at Flowable, stressed the importance of orchestration: of systems, and of roles. “It’s about doing agent orchestration the right way,” said Kiener, “with governance on top, knowing when to put the human in the loop, and when to put the model in the loop”.
From a business enablement perspective, leaders at Leena AI added that AI is now viewed as a “partner in operations.” Their team observed a shift away from one-off pilots toward integrated use cases that solve real business problems. “It’s becoming about being a business enabler instead of just an idea,” CEO Adit Jain, noted, a sentiment that reflects how AI maturity is now being measured in outcomes, when it has previously been measured in experimentation.
Jerry Walker, CIO for the Oregon State Treasury, shared how he’s turning that shift inward and using AI adoption as a catalyst for workforce development. “We’re not trying to eliminate jobs, but the work is changing,” he explained.
His teams are developing internal copilots, streamlining repetitive tasks, and even automating tone refinement for emails. “Something that used to take 15 minutes now takes 30 seconds, and I don’t even have to touch it,” he explained. “That’s where the real game-changer is”.
Strategy as a Team Sport
With technical execution on firmer ground, CIOs are turning their focus to strategy, and their roles are expanding accordingly.
“AI is only as good as the processes it runs on,” said Patrick Thompson, Global SVP at Celonis. He argued that without visibility into workflows, handoffs, and exceptions, CIOs risk building intelligence on top of dysfunction. Process Intelligence, he said, is the “connective tissue” to ensuring that AI delivers on its promises by connecting data to outcomes.
Prosymmetry CEO Sean Pales made a similar point through a human-capital lens: “Strategy execution starts and ends with your resource constraints.” As organizations chase transformation, many overlook the bottleneck hiding in plain sight: available talent and skills.
He advised starting with a clear picture of what your people can do, and matching that against your ambitions.
“You can’t govern what you can’t see.”
Wellington Estevao, Saviynt
That practicality also extended to customer-facing strategy.
Head of Machine Learning Solutions at Infor, Sandeep Anand, urged CIOs to “work backwards” from the business outcome. “When you want something done,” he said, “you don’t want a menu of ‘do you want half a thing?’ When you’re making a shirt, you don’t want to negotiate a sleeve.” Instead, he explained, the goal is to simplify technology delivery down to its essentials. “The customer wants one line item: I want to be able to do this. And the answer is yes or no.”
Looking Ahead
As organizations look toward 2026, the leading CIOs are chasing durability.
Aaron Miri, CIO of Baptist Health, anticipates a world of “sovereign data sets” where compliance and localization rules define what’s possible with AI. He also sees the role of hyperscalers shifting, as edge computing becomes a key strategy for optimizing performance and cost.
Meanwhile, Ardoq’s Bakstad expects a return to fundamentals: governance, visibility, and architectural clarity. “It’s going to come back to the fundamentals,” he said. “If you really want to get value from AI, you have to know what you have and bring the right context to it”.
Lakeside’s Klein put it even more succinctly:
Tal Klein, Lakeside
“AI should make people better, not redundant.”
The Wrap
If there’s a single thread tying these perspectives together, it’s clarity. Clarity of data, process, security, and purpose.
And with that clarity comes better decisions, stronger governance, and a higher return on every AI investment.
The organizations thriving in this AI-driven era aren’t necessarily those with the biggest models or the flashiest pilots, but they’re the ones with discipline. They’re making deliberate moves that prepare their organizations for what’s next, and the blueprint they’re following is built on governance, grit, and a clear understanding of what moves the enterprise forward
