AI’s opinion of the world’s opinion of you
Why AI recommends your competitor, and the three moves that change it
Since 2012 I’ve been telling brands that what they should care about is not what Google thinks of them and not what the world thinks of them, but the interpretation that sits between the two: AI’s opinion of the world’s opinion of you. The engine doesn’t invent its view out of thin air. It reads what the web already says about you, synthesises it, and hands the result to the person asking. So when that result is thin, hedged, or simply wrong, the cause isn’t the machine being unfair. It’s that the world’s view of you is scattered, contradictory, buried in corporate silos, and the machine has read exactly that and reported it accurately. Your job is to curate the world’s view until the machine reads it the way you’d want a fair, informed stranger to read it.
For a CMO or a CEO that reframes the whole problem, because it moves the question from “how do we beat the algorithm” to “what does the world currently say about us, and how do we make it clearer.” The second question has answers you can act on. The first never did.
The AI engine is an honest broker, and that puts the power back with you
The most useful way I’ve found to describe the engine to a board is as an honest broker. In business and diplomacy an honest broker is a respected, impartial intermediary: no hidden agenda, no stake in who wins, no price at which its recommendation can be bought. Its entire value to the person asking is that it can’t be leaned on. Think of any broker you’d trust, a financial adviser with the whole market on screen, an agent who carries every supplier in the category, and the AI engine behaves the same way. It carries every brand in your category, it answers to the buyer, and it recommends whoever it judges the best fit for them.
That impartiality is the reason it will recommend your competitor without a flicker of disloyalty: it was never on your side, it’s on the buyer’s. And here’s why that’s good news rather than bad. An honest broker has to stay honest, because the moment it can be rigged, buyers stop trusting it and leave for ChatGPT, Perplexity, or whoever stayed straight. So the engine genuinely wants to recommend the best answer, which means it wants to recommend you when you’re the right fit. It just can only recommend what it clearly understands and trusts. You don’t have to outwit a biased system. You have to give an honest broker the organised facts, more clearly and more often than your competitors give it theirs. Brief it better than they do, on merit, and you become the name it reaches for.
For me, that’s the shift that matters most for the people running brands today. The work isn’t manipulation and it isn’t a war against the machine. It’s briefing an impartial expert that’s already willing to speak well of you the instant you give it reason to.
Organise your digital footprint once, and every engine reads it
The objection I hear next in the boardroom is that this now means a separate strategy for Google, for ChatGPT, for Perplexity, for Copilot, for whatever launches next quarter, and that sounds like cost without end. It isn’t, and the reason is structural.
Every major AI platform runs on the same three underlying layers. There’s an index that reads and ranks the open web, a language model that does the reasoning, and a knowledge layer that stores verified facts and the relationships between them. Information, intelligence, verification: I’ve called these three the Algorithmic Trinity since 2024. The decisive point for a CMO is that all three layers, on every platform, pull from the same source, which is the open web. Your website, your reviews, your coverage, your profiles, the corroboration scattered across the market: that single body of material is what every engine reads.
So you’re not building one strategy for each engine. You’re cleaning up your corner of the internet so that every honest broker on earth reads the same pristine briefing. In our industry everyone has always optimised for Google first, because Google is the biggest by a long way, and the happy consequence in the AI era is that optimising for Google’s AI surfaces optimises for all the others at the same time. Organise the facts once, and you win everywhere the buyer might ask.
Your SEO investment carries straight into the AI era
CEOs are often told that the years and the budget poured into traditional SEO are now wasted, that AI has made all of it obsolete. That fear is worth taking off the table directly, because it’s wrong, and acting on it would mean dismantling exactly the capability the AI era rewards.
The interface changed at extraordinary speed, from a page of blue links to a conversational answer to an agent that buys on the user’s behalf. The mechanism a brand uses to claim its place inside that interface did not change. Optimising for AI is an extension of SEO, not a replacement for it: you’re still making your brand legible, credible, and easy for a machine to choose, with different tools and a wider set of surfaces. Better still, the work stacks in your favour. When you organise your data for the hardest case, the agent that just buys on command, you’ve automatically organised it for the assistive engine that recommends, and for the traditional search that lets the human decide. It’s fundamentally the same job, doing fundamentally the same thing, feeding the same facts to a machine that has more ways than ever to read them.
What to tell your teams
Three directives turn all of this into work that starts tomorrow morning: treat your footprint as training data, build for discovery rather than volume, and aim for the top of the broker’s mind. Each one is a brief to a different reflex your teams already have.
Treat your digital footprint as the training data it is. Your website, and every third-party reference to you, the reviews, the profiles, the citations, the coverage, is the material the engines learn you from. Audit it the way you’d audit anything you were handing to an examiner. Where it’s messy, inconsistent, or silent on something important, the broker can’t trust you, and an honest broker that can’t trust you stays quiet about you. Most of the value your business produces never makes it online at all, because it closed in a room or sat in a CRM, so the first task is bringing the true, detailed picture into the open where the machine can finally read it.
Build for discovery, not search volume. Stop measuring success by raw clicks and rankings alone. The engines now match a brand to the precise subset of buyers with a specific problem at the exact moment they need solving, so the work is to give the machine deep, structured, honest information about who you serve, what you solve, and the circumstances in which you’re the right answer. Volume is a metric from the era when the human did the searching. Discovery is the metric for the era when the broker does the matching.
Aim for the top of algorithmic mind. Frame your organic strategy as machine enablement rather than persuasion. The brand that makes its identity, its credibility, and its offering the easiest in the category for the honest broker to read and verify is the brand the broker reaches for first, every time, in front of buyers you’ll never meet. That position compounds: the earlier you build it and the more consistently you hold it, the harder it becomes for a competitor to displace, because the broker has been reading the same clear story about you for longer than it’s read anyone else’s.
None of this is a trick played on the machine. It’s the opposite. You’re showing an honest broker proof that was always true and simply never visible, and an honest broker, by definition, reads the better reality and passes it on. Change what the world can see about you, and you change what every engine says about you, at the one moment that pays: when a buyer is in the market for what you sell and the broker is about to be asked who it recommends.
