The Consensus Machine: Why AI Might Be Your Biggest Competitive Blindspot

I have written critically about AI before. And since then, it feels like the criticism has been growing.

Today I have more thoughts I felt I needed to share. Some of them may be unpopular. But they could mean the difference between actually moving forward as humans, leaders or businesses — or falling steadily behind.

So here it is: The AI-enabled world we are all rushing toward may actually be producing the opposite of what we're hoping for.

Not because AI isn't powerful. It is.

But because power, in the wrong direction, doesn't get you where you need to go faster.

It just gets you lost more efficiently.

It's been interesting to watch the leaders I work with over the past year. The ones most enthusiastically using AI for strategy, content, and decision support are producing work that is increasingly... similar.

To each other. To the industry. To the expected answer.

Polished. Coherent. Reasonable.

And yes increasingly…indistinct.

The rough edges that used to differentiate them — the idiosyncratic thinking, the counterintuitive gut instinct, the perspective that didn't fit the conventional narrative — are getting smoothed away. Replaced by something more optimized. More agreeable. More likely to be well-received.

And I've started asking: is this actually progress?

The Core Insight

Here is what AI is extraordinarily good at: giving you the answer you're most likely to find satisfying.

That is not a criticism. It is a description of how the technology works. AI is optimized for approval — for the confident, well-structured response that feels like the right answer, whether or not it actually is.

Which means when you ask AI a strategic question, you are not necessarily getting the truest answer.

You are getting the most agreeable one.

AI, at its core, is a consensus machine. It reflects the center of gravity of existing thought — synthesized and served back to you in a form designed to please. And the center of gravity of existing thought is, by definition, where everyone already is.

Which means if you are using AI to find your competitive edge, as a person or in business — you are using a tool optimized for consensus to find differentiation.

You won't need AI to tell you those two things are not compatible.

What AI Cannot Give You

There are things AI cannot produce. Not because it lacks processing power or data — but because they don't exist in the training set.

Your specific context. The particular combination of your market, your culture, your team, your history, your relationships — the texture of your specific situation that no generalized model can fully hold because they have grown slowly over time.

The tension that leads somewhere new. Real strategic insight is often uncomfortable. It challenges the existing direction. It surfaces the thing everyone has been avoiding. It requires sitting with disagreement and uncertainty long enough to find something genuinely new on the other side. AI is not optimized for productive discomfort. It is optimized for resolution.

The contrarian instinct. The sense that something in the conventional wisdom doesn't add up. That the obvious answer is obvious for the wrong reasons. That the thing everyone is doing is precisely why you shouldn't. That instinct — the one that fires before the analysis begins — is human. It is not reproducible by a system trained on consensus.

The question nobody is asking. Innovation almost never comes from better answers to existing questions. It comes from someone asking a different question entirely — one that disrupts the frame, challenges the assumption, opens a direction that the prevailing thinking hadn't considered. AI is exceptional at answering questions. It is structurally limited in its ability to question the question.

Your courage. The willingness to pursue a direction that the data doesn't fully support yet. To back a conviction that the consensus hasn't caught up to. To make a decision that will be judged harshly if it doesn't work — because you believe it's right. AI has no stake in the outcome. It cannot be courageous on your behalf. And the decisions that most require courage are often the ones that matter most.

The Courage Shift

Here is the uncomfortable implication of everything above:

If every one of your competitors has access to the same AI tools — trained on the same data, optimized for the same kind of approval, producing the same quality of agreeable, well-reasoned output — then AI fluency is not a competitive advantage.

It is a baseline.

The floor, not the ceiling.

And what differentiates the leaders and organizations that genuinely break through is not the quality of their AI outputs.

It is the quality of their human inputs.

The distinctiveness of their thinking. The courage of their convictions. The willingness to pursue the counterintuitive direction when the consensus says otherwise. The capacity to sit with tension and ambiguity long enough to find something genuinely new — rather than reaching for the comfortable answer that the machine is always ready to provide.

This requires a kind of courage that is becoming rarer, not more common, as AI adoption accelerates.

The courage to think before you prompt. To question before you generate. To sit with your own instincts — uncomfortable, half-formed, not yet supported by a beautifully structured output — long enough to find out whether they're worth pursuing.

The courage to ask not what is the best answer but what is the question nobody else is asking?

The courage to use AI as a tool in service of your thinking — rather than as a replacement for it.

Because the leaders who will genuinely differentiate in the next decade are not the ones who use AI most fluently.

They are the ones who know what AI cannot do — and who have developed the human capacities to fill that gap.

What This Means Practically

1. Use AI downstream, not upstream. Before you open the prompt, think. What do you actually believe about this problem? What is your instinct before the tool shapes it? What is the question underneath the question? Do your own thinking first — however rough and incomplete — and then use AI to pressure-test it, extend it, or stress-test the logic. The sequence matters enormously. Thinking before prompting produces something different — and often better — than prompting before thinking.

2. Ask AI for the argument against you. One of the most useful things you can do with AI is ask it to challenge your direction rather than support it. What are the strongest arguments against this strategy? What is this analysis missing? What would a smart critic say about this decision? Used this way, AI becomes a useful sparring partner rather than a comfortable validator. The discomfort it produces in this mode is a feature, not a bug.

3. Protect spaces for genuine disagreement. In a world where AI can produce instant, authoritative-sounding consensus, the meeting where people actually disagree — where tension is present, where the discomfort of divergent thinking is allowed to exist — becomes increasingly rare and increasingly valuable. Protect those spaces. Resist the urge to resolve the tension too quickly. The friction is often where the real insight lives.

4. Invest in what AI cannot replicate. Empathy. Judgment. Courage. Contextual intuition. The capacity to read a room, a relationship, a moment — and respond to what is actually there rather than what the data predicts. These are the human capacities that become more valuable, not less, as AI handles more of the analytical work. They are not soft skills. They are the differentiating skills. And they require deliberate, ongoing investment.

5. Treat your instincts as data. The contrarian thought. The nagging sense that something doesn't add up. The direction that feels right before the analysis supports it. These are not noise to be filtered out by better data. They are signal — the product of years of experience, pattern recognition, and contextual knowledge that no model fully holds. Develop the habit of surfacing and examining your instincts before outsourcing the thinking. They are often ahead of the consensus for a reason.

6. Ask the question nobody else is asking. Regularly. Deliberately. In your strategy sessions, your team meetings, your individual reflection time. What is the assumption everyone in this room is making that might be wrong? What is the direction our industry is moving that we should be moving away from? What would we do if we started from scratch — with no legacy, no sunk cost, no consensus to protect? These questions are not comfortable. They are not the kind of questions AI will generate unprompted. But they are where genuine differentiation begins.

A Note on What AI Is Actually For

None of this is an argument against using AI. It is an argument for using it more deliberately.

AI is an extraordinary tool for acceleration, synthesis, and execution. For processing information at scale. For removing friction from the parts of work that don't require human judgment. For freeing up the time and cognitive capacity that can then be invested in the thinking that actually matters.

Used well, it is a multiplier of human capability.

Used lazily — as a replacement for thinking, a shortcut to consensus, a way of outsourcing the discomfort of genuine strategic uncertainty — it is something else entirely.

It is a very efficient way of producing the expected answer.

And the expected answer, by definition, is already available to everyone.

What is not available to everyone is your specific perspective. Your particular combination of experience, instinct, context, and courage. Your willingness to ask the question that the consensus isn't asking — and to pursue the answer even when it's uncomfortable.

That is not something any model can generate.

It has to come from you.

A Courage Challenge

This week, before you open a single AI tool for a strategic decision — stop.

Write down what you actually think. Before the tool shapes it. Before the output makes the comfortable answer look like the right one.

What is your instinct? What is the tension you've been avoiding? What is the question that keeps surfacing that nobody on your team has been willing to ask out loud?

Write it down. Sit with it. Let it be rough and incomplete and uncomfortable.

Then — and only then — use AI to pressure-test it.

Ask it to argue against you. To find the holes. To tell you what the consensus says — so you can decide whether the consensus is right.

Because the leaders who will matter most in the next decade are not the ones who prompted their way to the best answer.

They are the ones who had the courage to think before they prompted.

To trust their instincts before the machine smoothed them away.

To ask the question nobody else was asking — and to pursue the answer, even when it carried tension.

Even when it wasn't what they wanted to hear.

Even when the consensus pointed the other way.

That is where your differentiation lives.

Not in the output.

In you.