Google wins consumer AI on distribution

Two billion people a month use a frontier AI model they never chose. No download, no sign-up, no new habit. Google put a Gemini-written answer at the top of the search results they were going to read anyway, and the choice got made for them with a server-side change. That’s worth more than it looks, and the reason starts with the least interesting part of it: the model.

Intelligence is the part that commoditises

The frontier labs are converging and the lead is now measured in months. China is the clearest tell. DeepSeek’s V4, released in April 2026, lands within three to six months of GPT-5.4 and Gemini 3.1 Pro, beats every open model on maths and coding, and trails only a closed model on world knowledge. It got there largely by distilling the frontier it’s chasing, training a cheap model on the expensive ones’ outputs. The accusation that this is industrial-scale copying is true and nearly beside the point. Distillation works, the floor keeps rising, and the ceiling gets matched a quarter or two after it’s set.

Intelligence is going the way every digital capability goes once several funded teams chase the same target, which is towards good-enough and cheap. That doesn’t make models worthless. It makes them a bad place to build a moat. If your plan is to win consumer AI by holding the smartest model, your position resets every few months against competitors who can rent or distil most of the advantage away. So the market gets decided by whatever doesn’t commoditise.

Distribution doesn’t commoditise

What doesn’t commoditise is the thing standing between a model and a person: the surface it lives on, and the habit of reaching for it. That was always the hard part of consumer software. Building something good is achievable; getting a human to form the habit of opening it is the expensive bit. OpenAI did exactly that from nothing, which is a real achievement, and ChatGPT is one of the fastest habits the consumer internet has formed, with 900 million people opening it weekly. It had to build that distribution a download at a time. Google built none. It already owns the surfaces where billions of people start the day, so the search bar, the browser, the phone, the inbox, the map and the video, and it can put a model on all of them with a configuration change rather than a marketing budget.

This is where AI as the interface stops being an abstraction. If the durable job of AI is being the layer you speak to instead of the system you learn, then the company that already owns the surfaces people speak into has the shortest path from a new model to a billion users of it. Apple holds the matching distribution, the other half of the world’s phones, and no frontier model of its own. So little of one that it’s now paying Google around a billion dollars a year for a custom Gemini to run the rebuilt Siri, white-labelled so the user only ever sees Siri, shipping to roughly 1.5 billion devices. Google’s model is about to power the assistant on both of the platforms people actually carry, its own and its only rival’s.

Neither does context

There’s a second moat hiding behind the first. Once intelligence is a commodity, usefulness stops being about how smart the model is and becomes about how much it knows about you. A brilliant model with no context gives you a brilliant generic answer. A fair model that knows your calendar, your inbox, your last ten searches, where you drove this morning and what you watched last night gives you the answer you actually wanted. Usefulness is intelligence times context, and the second term is where the contest moves once the first one flattens.

Nobody has more context on more people than Google. Gmail, Calendar, Maps with your location history, Search with everything you’ve ever asked, YouTube with everything you’ve watched, Photos with your life in it, Drive and Docs with your work, Android in your pocket, Home on the kitchen bench. That’s the most complete picture of a person any company has assembled, and it’s exactly the raw material a commodity model needs in order to stop being generic. The assistant-everywhere position feeds it further, because powering Siri and Gemini across both platforms is a firehose of real-world use and training signal. Even if Apple builds its own model and pulls the rug in a year, Google spends that year compounding its lead in the input that matters most. The model can be swapped out. The years of context it was tuned against can’t be handed back.

Consumers won’t pay, and Google doesn’t need them to

Distribution and context would matter less if consumer AI were a good business to be in directly. It isn’t. Consumers are famously bad at paying for software, the revealed preference of the median user is free with a hard ceiling on what they’ll convert to a subscription, and every query still costs real money to serve. That’s the quiet bind under the subscription AI companies. Inference is a marginal cost on every use, most users never pay, and the product they’re selling is the exact thing commoditising underneath them. Selling a melting asset to people who don’t like paying is a hard place to build anything durable.

Google is in a different business. It sells attention and monetises that attention through ads, and it has never sold intelligence. So Gemini doesn’t have to make money. Its job is keeping the customer inside Google’s surfaces long enough for the machine that does make money to run. AI as the interface to all of Google’s other products means the model isn’t the thing being monetised, it’s the funnel into the things that are. Giving intelligence away free is the rational move for the one company that profits from the attention rather than from the tokens.

Which is why commoditised intelligence, the thing threatening the subscription players, is a tailwind for Google specifically. As the price of intelligence falls towards zero, the company hurt most is the one whose product was the intelligence, and the company helped most is the one that wanted to give it away regardless. The same trend is a headwind for one business model and a subsidy for the other.

The same AI pays for itself twice

It gets better for Google than the AI holding the attention, because the same AI also monetises that attention harder. The fear for years was that an AI answer kills the search results page: resolve the question and you’ve removed the ten blue links and the ads stacked above them. Google’s answer is to rebuild the ad unit inside the answer. It’s testing conversational ad formats and AI-powered shopping ads directly in AI Mode, with Gemini writing the ad creative to fit the specific question, on a surface Google says has crossed a billion monthly users. The AI answer becomes new ad inventory rather than the end of the old inventory. Whether the yield per query matches the old page is the open question and I’ll come back to it.

The other half is ranking, and here Google is following Meta’s lead. Meta has spent the last couple of years pointing frontier-scale models at a narrower problem than chat: which ad to show, to whom, right now. Its Adaptive Ranking Model, live across Instagram in 2026, reads far more signals, including what people do with Meta’s own AI, to match an ad to the person most likely to act on it, and the lift is real. Google is doing the same across its surfaces with AI Max in Search, and on YouTube the kind of personalisation that used to be close to impossible, where a model reads both the viewer and the video they’re watching to place the ad that fits both. So the AI spend returns at both ends of the same pipe: as the interface holding the attention coming in, and as the ranking engine wringing more out of it going out.

Five fronts, one valuation

All of this rests on something rarer than any single product. Google is the only company holding every layer of the stack at once. The frontier lab is its own, DeepMind. The silicon is its own, the TPU, so it trains and serves without paying the Nvidia tax or waiting in the Nvidia queue. The cloud is its own. The models are its own. The surfaces that carry them to people are its own. That vertical integration is what lets it give intelligence away and still profit, because it captures the value downstream on infrastructure it owns end to end rather than renting a layer from a competitor.

The breadth is easy to under-feel because it’s smeared across so many products. Search, the most-used browser, the most-used mobile operating system, the most-used video platform, billion-user mail and maps and photos, the documents and drives where people keep their work, the speaker on the kitchen bench, the third-largest cloud, the leading robotaxi business, its own AI silicon, a frontier lab. Name a layer of the AI stack or a consumer surface that matters and Google is first, second or third in it.

The strange part is what that adds up to. Each company Google competes with fights on one front. Nvidia sells the silicon. Apple sells the devices. Amazon sells the cloud. Microsoft sells the enterprise seat. Meta sells the same thing Google does, attention monetised by ads, and is the one rival on exactly the front AI touches hardest. Google competes with all five at once, on each of their home grounds, and is a top-three player in every one of those markets. And it trades in the same band as any single one of them: Alphabet crossed four trillion dollars in January 2026 and passed Apple to sit second only to Nvidia, around 4.6 trillion against Nvidia’s 5.2, Apple’s 4.5 and Microsoft’s 3.1. The market spent years pricing it at a discount on the fear that AI would eat search. The re-rating since is a slow recognition that the company best placed to own the AI replacing search is the one that already owns search.

The case against

There’s a real case against all of this, and it deserves its strongest form.

Start with the cannibalisation the last section waved past. Even with ads rebuilt inside the AI answer, one resolved answer exposes far less monetisable surface than a page of ten links with four ads stacked above them. Google is betting it can reconstruct equivalent yield on a sparser surface, and that bet isn’t won. The search results page is the most profitable real estate ever built, and Google has to rebuild it in flight, doing to itself the thing a competitor would need years to do to it. Get the new yield wrong and the most reliable profit engine in technology degrades on purpose.

Reach also isn’t preference. The two billion are largely passive, served a summary they didn’t ask for, while the deliberate, high-intent relationship, the assistant you open on purpose to do real work, is where ChatGPT leads, along with revenue per user. If the valuable AI habit turns out to be the deliberate one, then impressions matter less than engagement and Google’s headline number flatters its position. OpenAI is building its own distribution too, through devices and the app habit, so the gap that looks decisive today might not stay this wide.

Commoditised intelligence cuts both ways as well. If the model edge erodes for everyone, it erodes for Google, and what’s left, distribution plus an ad machine, is the one thing Meta also has, with three billion users and an ad system of its own. The attention war might be a two-incumbent grind rather than a Google walkover. And the lever that makes Google’s distribution look unbeatable is the lever regulators most want to pull. The default-search deal, the Android bundle and the ad-tech stack are all live antitrust exposure, and a forced unwind of any of them would blunt the sharpest edge in the whole argument.

The most immediate problem is Google’s own. Gemini still trips on things it shouldn’t, and the company’s famously fragmented product structure shows up in experiences that don’t talk to each other. The context it holds is scattered across teams and apps that were never built to share it, which is half of why the assistant doesn’t yet feel like it knows you as well as it obviously could. That’s an execution problem rather than a structural one. The assets are all there, pointed slightly the wrong way, which you fix with org will and a few quarters rather than by going and acquiring something you don’t own.

Weigh all of it and the case still lands. The bet is that the best model stops being the thing that matters. Intelligence is commoditising, distribution and context aren’t, and consumer AI goes to whoever already has the attention, knows the most about the people they’re serving, profits from giving the intelligence away, and can turn the same AI into a sharper way to monetise that attention. Google is the only company with all of it, on the only stack that owns every layer.

Once every model is smarter than us at almost everything, owning the smartest one stops being an advantage. Knowing the user is the advantage that’s left.