Innovation Is The New Measure Of Digital Transformation

Jo Debecker is President and CEO of Akkodis.

getty

​This is the year of less is more.

Organizations are going to do less with AI, but they’re going to have to go all in on the AI they do pursue. AI can no longer sit on the edge of the enterprise as a productivity tool or a series of pilots. In the intelligence age, it has to become a real, active participant in how work gets done.

The next phase of AI requires CTOs to rethink the value of digital transformation itself, stepping away from efficiency and toward an operating model built around innovation, human-machine collaboration and AI agents.

CTO Confidence Isn’t Falling Because The Technology Failed

“What CTOs Think” research, featuring insights from 500 Chief Technology Officers, shows that CTO confidence in scaling AI is declining, falling to 48% in 2026 from 82% in 2024. But CTO confidence is not really falling in the capability or capacity of what AI can do. Most CTOs understand that this technology is powerful. Where confidence is falling is in the question of whether AI can actually scale in a business context.

That is where the challenge becomes bigger and broader than many leaders anticipated. AI is moving from a technology transformation program to an organization and business transformation program, which means the CTO and CHRO need to start talking.

Successful AI implementations looked beyond AI as only a productivity or automation tool to see that it can fundamentally change the way the business runs.

The Old Definition Of Digital Transformation Is Too Narrow

Many companies today are stuck in what I would call “pilot mode.” They can prove that a technology works in one place, but they struggle to adapt their operating model quickly enough to capture the full potential of AI.

Companies can claim to be more flexible, adaptive or efficient. But can they say their programs are driving innovation and creating business value?

If you want to automate or improve efficiency, use standard automation and deterministic technologies. You do not always need AI for that. But if you want to transform a business process, create business value and redesign how work flows across the enterprise, that is when AI comes into play.

The measure of AI success cannot be how many tokens have been used. Tokenization as a measure of AI success is passé now.

From Task Automation To Enterprise Orchestration

As defined in our latest research, organizations tend to fall into three archetypes: task automators, pilot operators and enterprise orchestrators. Task automators use AI primarily to optimize existing processes without fundamentally changing how work is structured, and in many cases, I would question whether AI is even the right technology for that goal. Pilot operators are experimenting and seeing their first business successes but have not solved for scale. Integration across strategy systems and workforce capability remains uneven.

Enterprise orchestrators are rethinking how work should be executed, where AI can take over tasks and how the hybrid workforce can work across humans and intelligent systems. AI is embedded into workflows, decision-making is shared between humans and machines and work is continuously redesigned.

It’s time to stop asking, “How can we automate this task?” and start asking, “How can we reduce time to market?”

Trust Is The Missing Link Between Pilots And Scale

The main issue in scaling AI is trust.​ Just 36% of CTOs are satisfied with workforce trust in organizational AI direction and decisions. When you deploy AI in real life, you touch humans. Humans are not perfect. Humans are messy. That is where trust becomes central.

Trust means giving people visibility into how AI is being used, where human-in-the-loop and central control points remain, how AI arrives at a result and what that means for responsibility and accountability.

What brought you successfully to the point of proving the value of AI will not make you successful in scaling and industrializing it.

With deterministic systems, trust is easier because you know how the machine will react. With agentic AI, you need a human control point because the technology itself is not deterministic. Building that control point at the right level gives trust back to the people who take final accountability, which is when acceptance and scaling become possible.

AI Agents Will Create A New Balance At Work

When a new member joins a family, the balance changes. The same is true when AI agents join a team. Leaders need to rethink how work gets done and how responsibility is distributed across intelligent systems, AI agents and humans. Hybrid workforce orchestration becomes a key differentiator.

Humans will still have the overall oversight and accountability. There is a reason some companies talk about copilots rather than autopilots and others talk about humans in the loop.

AI can do more than humans in certain domains. But humans can do more with AI than AI alone. That is why AI is fundamentally a human technology.

The future is finding the right balance between what humans do, what agents do and what they do together.

Leaders Need To Learn And Unlearn Simultaneously

The companies that use AI for real innovation will need curiosity. Combine this with decision-rights frameworks, accountability frameworks, operating models, governance structures, guidance and guardrails.

But leaders will also need to unlearn. They need to unlearn that digital transformation is all about cost and efficiency, that technology will replace humans instead of augmenting them and that current operating models are the only option.

Leaders also need to become more comfortable taking bigger steps while moving fast. It is okay to make mistakes, but make them fast because the space in front of us is larger than the space behind us.

AI Will Expand The World Of Work

I do not know of any general-purpose technology that has not created more jobs than it replaced, and I believe AI will move in the same direction.

When cars arrived, saddle makers were no longer jobs for life. But focusing only on that loss would have missed the car industry that came next: garages, roads, bridges, brand infrastructure. Software engineers and now even AI deployment engineers are all part of a new ecosystem of work.

We need to think in new dimensions. There are not enough horses on the planet to shoot a rocket into the sky, and leaders will not unlock AI’s potential by thinking only inside existing business models or today’s categories of work.

That is why innovation is becoming the new measure of digital transformation. To succeed, organizations will need to build trust, redesign work, partner across the organization and create operating models where humans and machines can do more together than either could do alone.​

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?