Lessons from Impact AI 2026: What Does Real Agentic Transformation Take?

CI&T’s fourth annual Impact AI brought together technology leaders and innovators at the New York Stock Exchange to explore one of the most pressing questions facing enterprises today: How do we move beyond AI experimentation to deliver meaningful, measurable business impact?

BEYOND THE HYPE: WHAT IT ACTUALLY TAKES TO SCALE

In the opening panel, moderated by Bruno Guicardi, the discussion tackled why so many enterprise AI initiatives remain stuck in pilot mode.

Most companies currently operate in augmented workflows equipped with copilots. While these tools yield 30% productivity gains, for many the math isn't closing because they’re seeking for marginal efficiency. True 20x performance leaps require re-engineering the core value chain, not just giving employees faster writing tools. 

Yet, the real fear for Fortune 500 enterprises is the operational reality of execution. As Bruno highlighted, capturing exponential value requires “deep-tissue transformation” but you cannot hit pause on current operations to build an ideal future state - a reality he describes as “changing the engine while flying.”

On the other hand, expecting employees to reinvent processes while simultaneously meeting quarterly goals creates friction, especially given that natural risk-takers represent only 10% to 15% of a company workforce.

THE WRONG MATH OF AI ROI

Why is the race to optimize token cost a fast way to get no return at all.

The point was never to make the token cheaper. It was to generate so much value that its cost stops being a question.

THE NEW ECONOMICS OF AI

Moderated by CI&T Partner and CMO Bob Wollheim, the second panel explored how AI is reshaping organizational cost structures, talent and process.

During the discussion, a provocation hung over the stage: To truly think new, remove the process. Then go further and remove the people. Following a pause in the room, came the twist: people are the only ones who can actually transform anything.

As a result, the dilemma is about the kind of room and environment we give people to drive reinvention.

While AI brings an abundance of intelligence, legacy and approval chains become the primary challenge to overcome. Simply generating output faster does not drive value if decisions remain trapped.

In his follow-up reflection, Bob emphasized that managing this shift requires embracing orchestration as a new management paradigm.

It operates on two distinct levels. Strategically, it aligns long-term outcomes with emerging capabilities. Operationally, it governs the daily interactions between humans and agents.

Ultimately, human excitement is the real catalyst. When enterprise leaders empower early-adopters and give them the space to reinvent how work gets done, curiosity spreads across the entire company.

GOING DEEPER THAN STAGE CONTENT

Between the stage discussions, over 60 enterprise executives gathered for roundtable sessions. Designed to facilitate candid exchanges, every table needed to leave behind a concrete answer to a specific question.

With AI opportunities outpacing enterprise budgets and capacity, leaders mapped how they decide which initiatives deserve funding.

Executives pointed where organizations must stop speeding up legacy tasks and instead completely rethink operations and processes.

As AI began recommending and making operational calls, leaders defined what should be autonomous and what must remain human.

Between the stage discussions, over 60 enterprise executives gathered for roundtable sessions. Designed to facilitate candid exchanges, every table needed to leave behind a concrete answer to a specific question.

AGENTS ARE COMING.
IS THE ENTERPRISE READY?

The final panel, featuring CI&T CEO and Founder Cesar Gon alongside Global Chief Business Strategy Paula Englert, addressed the shift from copilots to AI agents and Agentic SDLC.

As autonomous systems take on complex tasks across app modernization, data, and software delivery, moving beyond copilots requires orchestrated agent pipelines under human supervision to maintain governance at scale.

THE THREE WAVES OF AI DEPLOYMENT

Following the panel, Cesar reflected on how CI&T guides clients through this historical inflection point. Extracting value from AI requires three perspectives: capture efficiencies through optimization, select value streams to reinvent with AI at the center, and establish the governance needed to build trust and move fast.

He views this era as the most transformative in computer science history and this conviction keeps him leading a global team of 8,000+ AI builders into its next age.

BE PASSIONATE ABOUT THE PROBLEM, NOT SAFE WITH THE PROCESS

Looking one step ahead at the market dynamics, Paula Englert, Global Chief Business Strategy at CI&T, highlighted that business transformation starts by understanding human behavior and changing customer expectations.

As generations raised inside AI-native environments enter the workforce, technology acts as an enabler and is deeply connected to human demand. In this new landscape, value lives in solving problems and we need to overcome our psychological attachment to processes.

NAVIGATING WHAT’S NEXT

As we finished another edition of Impact AI at the NYSE, our conversations with enterprise leaders made clear that turning AI into impact is fundamentally a business transformation challenge, not a technology one.

We created Impact AI to give leaders a transparent space to share learnings and build actionable roadmaps together. We want to push these boundaries even further, exploring the limits of agentic orchestration and AI deployment to continue to help the largest enterprises in the world navigate change.

READY TO TAKE THE NEXT STEP ON YOUR TRANSFORMATION JOURNEY?

Download our newest e-book, Organizational Hallucination, to learn how your team can break free from pilots and start to deliver real impact.

---