AI Is Not a Strategy: It's an Amplifier. How to Build With It, Not Around It

There is a conversation I keep having. In boardrooms, in workshops, on video calls with teams across different industries and continents. It goes something like this. Someone senior asks what the team's AI strategy is. Everyone nods, as if this is an entirely reasonable question. And I feel the familiar itch to say: that is not quite the right question.

Not because AI is not important. It obviously is. But because calling AI a strategy is like saying your electric power tools are your building strategy. You do not need a cordless drill strategy. You need to build something. And you need to build it well. The drill is just one of several things that might help you do that faster.

Let me be clear, because I know how this can sound. At enterprise level, an AI North Star directive makes genuine sense. If you are a CEO or a board defining how AI reshapes your business model, your competitive positioning, your workforce over the next five years, that deserves a named strategy and serious investment. That is a different and necessary conversation.

But at team and org level, which is where most of us actually operate day to day, reaching for "AI strategy" as an answer is often a way of avoiding a harder, more honest set of questions. What problem are we actually solving? What outcome are we after? Are our fundamentals strong enough to amplify? And if they are not, will AI fix that, or just surface the chaos faster?

The Tool Confusion

The analogy I keep coming back to is this. Imagine you want to build a garden shed. You have a vision. You have the materials. But the project is moving slowly and the team is frustrated. So someone proposes: what we need is an electric powertool strategy.

The room gets excited. There are slides. There are pilot programmes for cordless drills. Someone runs a proof of concept on a circular saw. Months pass.

The shed is still not built.

Not because the power tools are bad. They are excellent. But the problem was never the tools. The problem was not having a clear plan for what was being built, why, and in what order. The shed has many components, needs multiple different tools, and requires a sequence. Power tools do not fix a planning problem. They let you make the same mistakes faster and at greater cost.

This is where a significant portion of organisations find themselves right now with AI.

What the Numbers Are Actually Saying

The data is not subtle. BCG's 2024 AI adoption research found that 74% of companies are struggling to achieve and scale value from their investments. McKinsey's State of AI report paints a similar picture: while 78% of organisations now use generative AI in at least one business function, most have yet to see meaningful bottom-line impact. More striking still, the proportion of executives reporting AI implemented with genuine, measurable impact declined from 37% in 2022 to 10% in 2024.

Read that again. Adoption went up. Impact went down.

This is not a technology failure. The technology works. It is a strategy and execution failure, and it is happening for a consistent, predictable reason: organisations are treating the tool as the answer when the question itself has not been properly defined. They are optimising the instrument before they have agreed on the music.

Less than 30% of companies have their CEO directly sponsoring their AI agenda. That matters, but not because CEO sponsorship magically delivers results. It matters because it signals whether the organisation has clarity on what it is actually trying to do. Without that clarity, you are not implementing a strategy. You are running an expensive and directionless experiment.

AI Amplifies. That Is the Point.

The organisations genuinely winning with AI share a consistent characteristic. They were already strong in the areas where they are applying it. They had clean data. They had clear processes. They had disciplined decision-making and a shared understanding of what success looked like. AI became the accelerant, not the foundation.

This is the part that often gets lost in the excitement. AI will surface what is already there. If your customer data is messy, AI will find patterns in the mess and present them with great confidence. If your internal processes are unclear, automation will lock that confusion into code and run it at scale. Garbage in, garbage out, but now at speed and volume.

I want to be clear on something, because I have seen this nuance get lost. Does AI also have genuine potential to replace and transform certain functions? Absolutely. In specific domains, workflows are being restructured in meaningful ways and things that once took weeks now take hours. That is real and worth planning for.

But for most teams, in most organisations, right now, the dominant near-term value story is amplification, not replacement. It is closing the gap between knowing what needs to happen and actually making it happen. Speed, quality, scale. That is the genuine unlock. And it only works if you know what you are trying to achieve before you pick up the tool.

A Note From Someone Who Used to Build Things

I started as an engineer. I genuinely love that building has never been easier. The fact that someone with a clear idea can now put together a working product in days rather than months is one of the more interesting shifts of our time, and I say that without irony.

But here is what years on the product side have taught me: building something and building a product are not the same thing.

A product is a thing that solves a real problem for a real person, consistently, at scale, in a way that can be managed, iterated on, and grown over time. That requires a specific kind of thinking. It requires understanding your user deeply, knowing the constraints, making hard tradeoffs and then living with the consequences. It requires thinking 360 degrees about how this thing behaves in the hands of real people in messy, real-world conditions.

Not everyone who can use a large language model or a no-code tool to produce something can think in that way. Effective product design, development, and management at scale is still an art. Change management across a global business is still one of the hardest things a leader can take on. The best product people I have worked with are not the fastest builders. They are the ones who ask the most inconvenient questions before anyone writes a single line of code.

AI makes the building faster. It does not make the thinking easier. That gap is where the real work still lives, and it is where competitive advantage is being built or lost right now.