AI Gave Everyone the Same Answers. The Advantage Moved to Better Questions.

Here's an uncomfortable fact about your marketing AI: your competitors have the same one.

Not a similar one. The same one. The same models, the same training data, the same optimisation logic. When you ask Claude to write ad copy for a plumbing business in Adelaide, and your competitor three suburbs over asks the same question, you get functionally identical output. Same headlines. Same calls to action. Same landing page structure. Same "professional, reliable, trusted" positioning that means nothing because everyone claims it.

When a tool is rare, it's an advantage. When everyone has it, it's infrastructure. Electricity was a competitive advantage in 1900. By 1950, it was just the cost of doing business. AI marketing tools crossed that threshold faster than anyone expected.

The Logic Trap: When Every Calculator Gets the Same Answer

Rory Sutherland, Vice Chairman of Ogilvy UK, has a line that should be tattooed on every marketer's forearm: "It doesn't pay to be logical if everyone else is being logical."

His argument, laid out in Alchemy, is that logic gets you to exactly the same place as your competitors. If you're optimising the same variables with the same data using the same frameworks, you arrive at the same answer. And sharing a market position with everyone else is not a profitable place to be.

AI is the most powerful logic engine ever built. It's also the most widely distributed one. Which means it's the fastest path to competitive convergence in marketing history.

Think about what happens when every business in your category feeds the same inputs into the same AI:

A February 2026 Forbes analysis described the mechanism precisely: generative AI models are, by design, pattern-matching machines that identify what has been done and produce variations of it. They regress toward the most statistically common forms of expression, reasoning, and problem-solving. The bold ideas, the contrarian perspectives, the distinctive brand voices are eliminated and replaced with consensus thinking that AI indexes. What remains is the algorithmic middle: safe, templated, and entirely forgettable.

This is Sutherland's logic trap, automated at scale.

The Sea of Sameness Has a Name: Regression to the Mean

Byron Sharp and Jenni Romaniuk at the Ehrenberg-Bass Institute have spent decades studying what actually makes brands grow. Their research across 1,162 distinctive brand assets, 21 categories, four countries, and nine years points to one consistent finding: brands grow by being distinctive, not differentiated.

Differentiation says "we're better." Distinctiveness says "you'll recognise us instantly." The Coca-Cola contour bottle. The McDonald's golden arches. Bunnings' green and red. These aren't arguments for superiority. They're memory shortcuts that trigger recognition in buying situations.

Sharp calls this mental availability: the probability that a buyer thinks of your brand when a purchase occasion arises. It's built through consistent, distinctive brand assets deployed with broad reach over time.

AI works against this in two ways.

First, it produces output that converges toward the mean. Every brand drafts with the same models, the models pull toward the same statistical average, and the category converges until you can't tell one business's website from another's. Forbes reported in September 2026 that this "sea of sameness" is now the single biggest risk in marketing communications.

Second, AI optimises for what has worked before. But distinctiveness, by definition, requires doing something that hasn't been done. You can't build a distinctive brand asset by copying the statistical average of all existing brand assets. That's how you build an invisible one.

Sharp's recent research shows that shape-based assets (logos, packaging, visual devices) achieve 40% Fame and 71% Uniqueness, the highest of any asset type. These are precisely the elements that require deliberate human creative decisions. They can't be generated by asking AI "make me a logo that works."

We've written before about why AI can generate a thousand ads but still can't build your brand. The research is getting clearer on why: brand building requires the courage to be distinctive, and AI is architecturally designed to avoid that.

The Centaur Principle: Why Process Beats Power

In 1997, Garry Kasparov lost to IBM's Deep Blue and chess changed forever. But the most interesting thing that happened next wasn't computers dominating humans. It was what happened when they started working together.

Kasparov invented "Advanced Chess" (also called centaur chess): humans collaborating with chess engines as a team. In 2005, a PAL/CSS Freestyle Tournament produced a result that startled the chess world. The winners weren't grandmasters with supercomputers. They were two amateurs, Steven Cramton and Zackary Stephen, with chess ratings between 1,400 and 1,700 (well below expert level), using three ordinary chess programs.

They beat grandmasters with better computers.

Kasparov's explanation has become one of the most cited observations in human-AI collaboration: "Weak human + machine + better process was superior to a strong computer alone and, more remarkably, superior to a strong human + machine + inferior process."

The quality of the human's judgment about when to listen to the machine and when to override it mattered more than the power of either the human or the machine individually.

This maps directly onto marketing. The business that wins isn't the one with the most sophisticated AI tools. It's the one with the best process for deciding:

• What questions to ask (strategy and diagnosis)
• Which AI outputs to use and which to reject (taste and judgment)
• When to do something the AI would never recommend (courage and distinctiveness)

That's why we've argued that the real opportunity in AI wasn't cost reduction. The real opportunity is freeing up human time for the diagnostic and strategic work that AI can't do.

What AI Can't Do: Diagnose, Decide, and Dare

Taste isn't subjective preference. It's informed judgment built through deep experience with what actually works. It's the Glossier social media team spending weeks answering customer support emails before writing a single post, so they understood how customers actually talk about their problems. It's knowing that the statistically optimal ad headline might be the worst choice for your specific brand because it sounds exactly like everyone else.

This connects to Binet and Field's effectiveness research from the IPA Databank. Their analysis of nearly 1,000 campaigns found that brands combining long-term brand building with short-term activation see +90% average ROI uplift compared to performance-only approaches. Brand building, the long game, the distinctive creative, the broad-reach campaigns that don't optimise for clicks, produces the compounding returns that separate growing businesses from stagnant ones.

And brand building is precisely the thing AI is worst at. Because brand building requires:

The businesses pulling ahead in 2026 aren't the ones using AI the most. They're the ones using it most deliberately, combining AI execution with human judgment about three things AI can't provide:

What This Means for Your Business

If you're an SME spending $2,000 to $10,000 a month on digital marketing, here's the practical takeaway.

The irony of the AI revolution in marketing is this: the more powerful the tools become, the more valuable the human decisions become. AI commoditised execution. That made strategy the last remaining competitive advantage.

Your competitors have the same AI you do. The question is whether you're asking it the same questions they are.

Further Reading

• Taste Is a Competitive Advantage - Sam Tomlinson on why human curation separates winners in an AI-saturated market
• Why AI Is Drowning Your Marketing Into a Sea of Sameness - Forbes on the homogenisation problem
• The Real Risk of AI Is Marketing Commoditization - MarTech on AI's convergence effect
• Shape-Based Assets Are Strongest: Benchmarking Distinctive Brand Assets - Ehrenberg-Bass Institute research on what makes brands memorable
• Advanced Chess - The centaur chess experiments that proved process beats power