AI is giving everyone access to increasingly similar capabilities. Execution becomes the competitive advantage.

If AI is giving everyone access to increasingly similar capabilities, what becomes your competitive advantage?

My view is simple. Execution.

AI increases potential. Execution determines outcomes.

For decades, knowledge created advantage

Businesses have always competed through access. Access to information. Access to specialists. Access to expensive research. Access to technology.

If your organisation knew something others didn't, you had an edge. Knowledge itself was a competitive moat.

AI changes the economics of knowledge

Large language models have dramatically reduced the cost of the first draft. A marketing strategy. A project plan. A business case. An explanation of an unfamiliar topic. AI can produce a competent first version of all of them in minutes.

And this is not opinion. It has been measured. In a landmark field experiment, researchers from Harvard Business School and BCG gave 758 consultants access to GPT-4. Those using AI completed 12.2% more tasks, worked 25.1% faster, and produced work that was rated more than 40% higher in quality.

Here is the part that should stop every leader. The lowest-performing consultants improved the most. On many of the tasks studied, AI reduced performance differences by lifting the performance of those who previously struggled the most.

Competent output has become easier to produce. That means competent output is no longer the competitive advantage.

Some will say execution is next

I know the counterargument. If AI came for knowledge work, agents will come for execution. Scheduling. Follow-through. Task completion.

Partly true. Agents will increasingly complete tasks. But execution is not task completion.

Execution is alignment. Prioritisation. Accountability. Deciding which of fifty ideas deserves the next six months, then having the discipline to stop the other forty-nine.

Judgement chooses the right thing. Execution delivers it, repeatedly, across teams and over time. No AI system consistently owns that today. The organisations that do will pull ahead.

The pattern is already visible

One widely discussed report from MIT's NANDA project, The GenAI Divide: State of AI in Business 2025, found that despite an estimated USD 30 to 40 billion invested in generative AI, 95% of pilots showed no measurable P&L impact within the study window.

The critical finding was why. Not the models. The report pointed instead to challenges around organisational integration, operating models and prioritisation. In other words, execution problems.

The report's six-month measurement window has been debated, and longer-term outcomes may differ. Even so, it reinforces a broader pattern seen across organisations: AI adoption is accelerating far faster than organisational transformation.

BCG reaches a remarkably similar conclusion through a different lens. Its widely cited 10-20-70 rule argues that around 10% of the effort lies in algorithms, 20% in technology and data, and roughly 70% in people, processes and organisational change.

Read that again. Most of the work required to create value from AI happens inside the organisation, not inside the model.

Technology rarely fixes organisational problems. It reveals them first. Then it accelerates them.

Leaders should ask different questions

Many organisations are still asking which AI tool to buy. That is a reasonable question. It just isn't the most important one. The better questions are these:

• How quickly can we make decisions?
• How effectively do our teams align?
• How rapidly can ideas become products?
• How consistently do we deliver?

Those questions existed before AI. They have simply become far more important.

AI is becoming infrastructure

Nearly a decade ago, Andrew Ng described AI as "the new electricity". That comparison feels increasingly accurate.

Nobody wins simply because they have electricity. They win because of what they build with it. AI is moving in the same direction. Organisations will not be judged by whether they use AI. They will be judged by what they accomplish because they use it.

The bottom line

AI increases potential. Execution determines outcomes. AI is commoditising access to knowledge. Execution is becoming the competitive moat.

That may prove to be one of the defining business lessons of the AI era. The winners will not have the smartest AI. They will build the smartest organisations.

So here is the question I will leave you with. If your closest competitor woke up tomorrow with exactly the same AI capabilities as your organisation, what would still make you win?

---

Sources

Dell'Acqua, F., et al. Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality. Harvard Business School / BCG field experiment (2023), later published in Organization Science.

MIT NANDA (Project NANDA, MIT Media Lab). The GenAI Divide: State of AI in Business 2025.

BCG. The Leader's Guide to Transforming with AI (10-20-70 rule).

Andrew Ng. Stanford Graduate School of Business and AI keynote presentations (2016–2017).