AI: The competitive advantage of disciplined execution over speed

The window for experimentation is closing 

With €2.53 trillion in AI spending forecast for 2026*, organizations face a stark reality. Research shows that 42% of AI initiatives are abandoned before production**, and the cost of architectural missteps is rising.

If you ask me, AI succeeds when the groundwork is laid first. This blog builds on that thinking, drawing insights from Fujitsu’s AI mini-report*** developed with FT Longitude, and looks at why disciplined execution - not speed alone - has become the real competitive advantage.

AI is no longer operating in a regulatory grey area. Increasing scrutiny means organizations must think about compliance, governance, and risk from the outset—not after deployment.

Fernando Almeida, Fujitsu’s Head of Portfolio Strategy Hybrid IT, captures this shift clearly:

“Regulation is no longer something you deal with after deployment. It shapes how AI systems are designed from the start.”

This fundamentally changes how AI initiatives are approached. Speed alone is no longer a competitive advantage—execution quality is.

The foundation determines the ceiling 

As AI matures, a clear pattern is emerging: organizations that prioritize strong foundation—data, architecture, governance—outperform those that prioritize speed alone. That’s exactly what I have also described in my blog ‘Why AI fails without application modernization’*.

As Almeida explains:

“You need to set up the foundation first, and that part is slow, because you have to agree on architecture, governance, and operating models in advance.”

But that investment pays off:

“Once that foundation is in place, you can adopt new AI capabilities quickly and safely.”

At the core of this foundation is data. Without it, even the most advanced AI systems will struggle to deliver value.

Walsh emphasizes this point:

“Before you think about tools, you need to understand your data — where it is, who owns it, and what you are allowed to do with it. Otherwise, AI creates risk instead of value.”

Security starts with transparency

As AI adoption accelerates, so too does the threat landscape. In my earlier blog, ‘Don’t let AI drive without a seatbelt*’ I explored how new tools, interfaces, and dependencies are rapidly expanding the attack surface—making security a core requirement, not an optional layer. This shift calls for a fundamental rethink of the role of security teams. Rather than acting as gatekeepers, they must evolve into enablers of business change—embedding guardrails that allow innovation to move forward safely and at speed.

John Swanson, Fujitsu’s Global Security Portfolio Lead:

“Security teams can’t afford to be gatekeepers anymore. Their role is to enable safe innovation.”

At the same time, not all risks originate externally. In another recent blog, I discussed how internal behaviors—such as employees using unauthorized AI tools, often referred to as “shadow AI**”—can introduce significant governance challenges. Yet viewed through a different lens, this same phenomenon can also signal bottom-up innovation, highlighting unmet needs and opportunities for the organization to harness more effectively.

The use case imperative 

One of the biggest reasons AI initiatives fail is not technology—it’s a lack of clear purpose. Without defined outcomes, projects drift, stall, or are abandoned before reaching production.

Disciplined execution means focusing on meaningful use cases from the start, aligning stakeholders, and ensuring accountability across the business. AI success depends on solving real problems—not pursuing innovation for its own sake.

Trust is the AI adoption multiplier

Even with strong governance and security, AI cannot scale without trust. Employees need to understand how AI works, where its boundaries lie, and when human judgment is required.

Swanson highlights the human dimension:

“It's about understanding people, understanding AI in terms of how it can be used, what the guardrails are, and the ethics associated with it.”

Today, many organizations are still using AI in limited ways—often as a more powerful search tool. But the real opportunity lies in transforming how work gets done.

This shift requires careful balance. AI can accelerate decisions, but it cannot replace human accountability.

As Walsh cautions:

“AI is a highly valuable tool, but it’s not an oracle. ‘Probably right’ is simply not acceptable. When decisions involve judgement, ethics or accountability, humans have to remain in control.”

Foundations first: why disciplined execution beats speed

Successful AI adoption does not come from big-bang transformations or headline-grabbing pilots. It comes from controlled progress, clear ownership, and an honest understanding of what success looks like before scaling begins.

As Fernando Almeida concludes:

“If you try a big-bang approach with AI, it usually fails. Scaling only makes sense once you know what success looks like.”

That is the uncomfortable truth many organizations are now confronting. AI rewards those who invest early in foundations - governance, security, data, and trust - and punishes those who confuse motion with progress.

The real competitive advantage in AI will not belong to those who moved first. It will belong to those who built strongest. Organizations that treat AI as a long-term capability, not a short-term experiment, will be the ones best positioned to scale with confidence and deliver lasting business value.