AI is "Just Pattern Recognition"

I've heard this in boardrooms, strategy meetings, and many executive offsites.

It's delivered with a knowing nod, a slight smirk—the tone of someone who's seen through the hype and won’t be fooled by the AI evangelists.

And they’re right. LLMs are fundamentally about pattern recognition.

But here’s the uncomfortable question that dismissal reveals:

If pattern recognition is so basic, so trivial, so “just”—why is your organisation spectacularly terrible at it?

The Rearview Mirror Problem

The header image for this post is from the 1976 cult film The Gumball Rally. Before the start of the race, Franco, one of the drivers, rips the rearview mirror off his Ferrari and tosses it away, declaring:

It’s a great cinematic moment — pure confidence, pure speed. But it’s also the perfect metaphor for modern leadership.

In our obsession with “leading indicators” and forward momentum, many leaders have become Franco behind the wheel — looking only ahead, never back.

And in doing so, they’ve forgotten that progress without reflection isn’t strategy. It’s amnesia on wheels.

You can’t improve your ability to see what’s coming if you never study what you’ve already hit.

The Intelligence You’re Already Generating (And Ignoring)

Let me ask you something. Over the past three years, how many of these has your organisation produced?

• Quarterly OKR update meetings
• Mid-quarter business reviews
• End-of-year retrospectives
• End-of-quarter post-mortems
• Strategy offsites
• KPI dashboards and reports
• Major project update meetings
• All-hands presentations
• Board updates
• Customer feedback sessions

Dozens? Hundreds? Thousands of hours of discussion, analysis, and strategic reflection?

Now answer this: when was the last time you systematically analyzed the patterns across all of them?

• When did you last ask:
• Which initiatives consistently miss their targets, and what do they have in common?
• What warning signs appear in struggling projects three months before they officially fail?
• Which types of quarterly adjustments actually improve outcomes versus which ones just delay the inevitable?
• What patterns emerge in the gap between what leadership says matters and what actually drives results?
• Which assumptions show up in every annual plan and prove wrong every single year?

The silence in response to these questions isn’t because the data doesn’t exist.

It’s because you’ve never looked for the patterns.

Strategic Amnesia Meets Pattern Blindness

In my previous article on strategic amnesia, I noted that 85% of leadership teams spend less than one hour per month discussing strategy — and 80% of companies fail to track their strategic goals. But it’s actually worse than that.

You’re not just failing to review last year’s strategy.

You’re sitting on a goldmine of strategic intelligence — recordings, transcripts, dashboards, documents — and treating it like worthless noise.

Every quarter, you generate more data about how your organisation actually operates versus how you think it operates. Every retro reveals friction points. Every missed OKR tells a story. Every strategy offsite documents your assumptions about market, capability, and competition.

And every single cycle, you file it away and start fresh, as if you’re a goldfish encountering the castle in its bowl for the first time.

That’s not strategic planning.

That’s strategic amnesia — with extra steps.

The Pattern Recognition Test

Want to know if your organisation is any good at pattern recognition? Answer these questions:

1. Can you identify the common factors across your last five failed initiatives without checking any records? If not, you're about to fund the sixth one.
2. Do you know which executive's projects consistently deliver and which ones consistently miss—and what that tells you about capability allocation? If not, you're pattern-blind to your own bench strength.
3. Can you describe the typical trajectory of how priorities shift mid-year, and whether those shifts improve or degrade outcomes? If not, you're repeating behaviors you can't see.
4. Have you identified the specific phrases that appear in confident Q1 project pitches that predict Q4 disappointment? If not, you're funding based on presentation skills, not probability.
5. Do you know what percentage of your "this time will be different" initiatives actually turn out different? If not, you're trapped in a loop you refuse to acknowledge.

This is pattern recognition. Basic, fundamental, "just" pattern recognition. And you're probably not doing any of it.

The AI Paradox

So let's return to that dismissive comment: "LLMs are just pattern recognition."

Here's what makes this dismissal so revealing. The same executives who use "just pattern recognition" as a reason to ignore AI are leading organisations that:

• Make the same strategic mistakes repeatedly because they never identify the patterns
• Waste millions on initiatives that follow failure patterns they could easily spot
• Promote leaders based on pitch quality rather than delivery patterns
• Set goals using the same assumptions that failed last year, and the year before, and the year before that

You're not dismissing AI because pattern recognition is trivial. You're dismissing AI because accepting its value would force you to confront how comprehensively you've failed at something that basic.

What "Just Pattern Recognition" Could Actually Do

Imagine feeding every recorded strategy meeting, every OKR update, every project retro, every KPI dashboard from the past three years into a system designed for pattern recognition.

What would it reveal?

• The specific language that predicts project failure with 80% accuracy
• The common factors behind initiatives that beat their targets
• The gap between stated priorities and actual resource allocation
• The seasonal patterns in how teams adjust their goals and whether those adjustments work
• The early warning indicators that appear months before problems become crises
• The difference between what leadership thinks drives success and what actually does

This isn't science fiction. This is literally what pattern recognition does. It finds signals in noise. It identifies recurring themes across massive datasets that human memory and attention can't hold simultaneously.

You're sitting on strategic intelligence that could transform your decision-making, and you're actively choosing not to look at it.

The Real Threat Isn't AI

The companies that will leave you behind aren't the ones with better strategic plans. They're the ones that finally figured out how to close the learning loop.

They're taking all those meetings, all those reviews, all those retrospectives, and actually extracting the patterns. They're building institutional memory that compounds. They're identifying what works in their specific context, not what works in theory.

They're treating pattern recognition as the competitive advantage it actually is.

Meanwhile, you're in another strategy offsite, with another pristine whiteboard, making another annual plan that carefully ignores everything you learned last year because you never bothered to look for the patterns in the first place.

The Choice

You have two options:

Option 1: Continue dismissing pattern recognition as "just" something trivial. Keep generating thousands of hours of strategic discussion and analysis. Keep filing it away. Keep starting each planning cycle as if you've learned nothing. Keep wondering why your success rate on strategic initiatives hovers around 10-20%.

Option 2: Accept that if pattern recognition is so basic, you should probably be good at it. Feed your strategic history into systems designed for exactly this purpose. Learn what actually predicts success in your organization. Build compound strategic intelligence that makes each cycle smarter than the last.

The irony is exquisite. The same executives who dismiss AI for being "just pattern recognition" would kill for the strategic insights that pattern recognition would reveal about their own organizations.

You don't need better strategies. You need to actually learn from the strategies you've already tried.

The data exists. The technology exists. The only thing missing is the willingness to look at the patterns you've been ignoring.

So the next time someone says "LLMs are just pattern recognition," ask them this:

"That's interesting. What patterns have you recognized in our last three years of strategic reviews?"

The silence that follows will tell you everything you need to know about whether they're dismissing AI's capabilities or defending their own strategic blindness.