What Brand Strategy for AI Companies Can Learn From Apple
May 12, 2026
January 9, 2007. Steve Jobs walks onto a stage in San Francisco and tells the room he is about to introduce three new products.
A widescreen iPod with touch controls.
A revolutionary mobile phone.
A breakthrough internet communicator.
The audience applauds each one. Then Jobs pauses, lets the room settle, and says: these are not three separate devices. This is one device.
Nobody in that room had asked for it. The gap between that fact and how most brand strategy for AI companies operates is the subject of this piece. Nobody had written it on a survey, sketched it on a napkin, or described it in a focus group. There was no expressed demand for a touchscreen smartphone with no keyboard. The desire Jobs manufactured in those ninety seconds was brand-new. And yet, by June of that year, people were sleeping on pavements to be first in line.
The iPhone was not a response to consumer demand. It was a demand-creation event. And if you are working on brand strategy for AI companies right now, that distinction is the most important one you will encounter.
In this piece:
1. The Demand Capture Trap: Why Most Companies Play the Wrong Game
2. What Apple Had Before the iPhone: The Brand Architecture It Drew Against
3. The 2007 Keynote Was Not a Launch. It Was a Belief Event.
4. The Three Requirements of Demand Creation at the Extreme
5. Why Brand Strategy for AI Companies Requires the Same Logic
6. Frequently Asked Questions
7. Build the Altar Before You Bring the Offering
The Demand Capture Trap: Why Most Companies Play the Wrong Game
Most brand strategy is demand capture.
You identify where desire already exists, position your product to intercept it, and spend your budget making sure you are visible when the purchase decision happens. It is a rational strategy. For existing categories, it is often the right one.
The trap is applying demand capture logic to a product that belongs to a category that does not exist yet.
Al Ramadan, Dave Peterson, Christopher Lochhead, and Kevin Maney spent an entire book explaining why this is fatal. In Play Bigger (HarperBusiness, 2016), they introduced the discipline of category design: the idea that the most valuable companies do not compete for markets, they create them. Their data is unambiguous. Category kings, the companies that create and define a new space, capture approximately 76% of that category's economic value. The companies that follow capture the rest, divided across everyone else.
You cannot build a legendary company without building a legendary category.
Here is what makes the iPhone case so instructive. No market research in 2006 would have surfaced a touchscreen smartphone without a physical keyboard as a latent consumer need. This is not a hypothetical. Nokia and Motorola were conducting exactly this kind of research, speaking to exactly the right audiences, and concluding that no such demand existed. Clayton Christensen's Jobs-to-Be-Done framework, laid out in Competing Against Luck (HarperBusiness, 2016), explains the structural reason: consumers can only articulate desires for things within their existing frame of reference. They can tell you what is frustrating about the phone they have. They cannot tell you about the phone they have never imagined.
Demand capture requires desire that already exists. When the product is genuinely new, that desire has to be built from scratch. The only available strategy is demand creation.
And most brand strategy for AI companies today is still doing the wrong one.
What Apple Had Before the iPhone: The Brand Architecture It Drew Against
Yes, it is Apple. Stay with me.
The iPhone was not Apple’s first act.
By January 2007, Apple had spent a decade constructing a very specific kind of cultural credibility. The iMac in 1998 repositioned the company from near-bankrupt to aspirational. The iPod in 2001, launched with the line a thousand songs in your pocket, did not just sell music players. It rebuilt Apple's identity from the ground up.
Here is why that tagline matters. A thousand songs in your pocket is not a spec. It does not tell you the storage capacity, the battery life, or the file formats supported. It tells you how life should feel. It is a worldview statement, compressed into six words. According to Walter Isaacson's biography Steve Jobs (Simon & Schuster, 2011), Jobs invested $75 million of Apple's marketing budget into the iPod specifically because he believed it would reposition all of Apple. Not just sell units. Reposition a company.
He was right. By January 2007, iPod accounted for nearly half of Apple's revenues, and Apple had become the most aspirational consumer technology brand on earth.
This is what I would call a trust reservoir: a decade of brand decisions, each one making a deposit, each one earning the company a little more permission to ask something unreasonable of its audience. When Jobs walked onto the Macworld stage in 2007, he was not asking the room to trust an unknown product. He was asking them to extend trust they had already given. The iPhone drew against an account that had been filling since 1997.
There was also a second layer. Jobs himself had accumulated a form of cultural authority that functioned as brand architecture in its own right. Apple engineer Bud Tribble coined the term reality distortion field in 1981, documented by Isaacson, to describe Jobs's uncanny ability to make people believe his definition of what mattered. By 2007, when Jobs called something revolutionary, the word carried evidential weight. He had been right before. Repeatedly.
As Sue Halpern observed in The New York Review of Books in 2012, reviewing the Isaacson biography, Jobs had made himself effectively inseparable from the brand he built. The man was the architecture. And the architecture made the launch possible.
The 2007 Keynote Was Not a Launch. It Was a Belief Event.
Rewatch the 2007 keynote with a strategist's eye, not a tech enthusiast's.
Jobs opens not with features but with a claim: every once in a while, a revolutionary product comes along that changes everything. He then places the iPhone in explicit lineage alongside the Macintosh in 1984 and the iPod in 2001. Not as a product comparison. As an argument about what this moment means, calibrated against events the audience already understands as historically significant.
The three-products reveal was pure semiotic engineering. Three categories the audience already trusted and understood, collapsed into one object that did not yet exist in their mental models. The desire had to form in real time. And it did, because the brand architecture made it possible to transfer meaning from the known to the unknown.
A 2026 retrospective on Apple's product marketing strategy put it precisely: the keynote format became its own marketing instrument, one stripped of spec sheets and feature lists, built instead around a carefully constructed story designed to create a moment. Not a campaign mechanic. Brand architecture.
The audience desired the resolution before they knew the specifications.
The same principle appears in Apple's pre-launch secrecy. The iPhone project, internally codenamed the Purple Project, was so tightly guarded that Jobs reportedly plastered the offices with Fight Club-themed posters reminding employees of the first rule. The secrecy was not just IP protection. It was demand engineering. By the time of the January announcement, there were six months of speculative desire already stored up. The waitlist had not opened. The product had not been seen. The desire existed anyway, because Apple had built the architecture that made it possible.
The Three Requirements of Demand Creation at the Extreme
What Apple did in 2007 was not magic. It was a strategy with three specific preconditions. All three. Not two.
1. The trust reservoir must pre-exist the product
Demand creation requires that consumers already trust your judgment before the product is known. Apple had a decade of that trust built in. Google did not when it launched Glass in 2013. Google's brand equity at the consumer layer was functional: search utility, productivity, organisation. The desire Glass required was aspirational. The mismatch was structural. The trust reservoir was the wrong shape, not just the wrong size. Glass asked people to join a category nobody wanted to be seen joining, and Google had built no cultural credibility that would have made that ask reasonable.
2. Communicate worldview first, features second
A thousand songs in your pocket is not a spec. It is a statement about how life should feel. Apple never led with gigahertz. They led with what their products believed about the world. The Ramadan et al. framework in Play Bigger calls this the Point of View: the company's articulated position on what the future looks like and why it matters. For demand creation, you are not selling what the product does. You are selling the world that becomes possible when people have it.
Most brand strategy for AI companies today leads with capabilities: benchmark scores, context windows, integration counts. This is demand capture logic applied to a demand creation market. A user who has never used an AI agent has no frame for evaluating a MMLU score. The benchmark communicates to nobody outside the early adopter community. It is the 2007 equivalent of a competitor announcing their new touchscreen phone had a 3.2 megapixel camera. Technically differentiating. Contextually meaningless to the audience you need to move.
3. Build the desire before the product arrives
Desire is a pre-launch project, not a launch-week one. The cultural heat Apple generated between the January 2007 announcement and the June 2007 release was not accidental. Six months of anticipation, speculation, and narrative accrual before a single unit shipped. This is why the sequencing of brand investment matters so much. The waitlist is not a growth tactic. Handled correctly, it is a demand architecture instrument.
Why Brand Strategy for AI Companies Requires the Same Logic
Here is the uncomfortable parallel.
AI companies are building products people have never used in the forms being launched: AI agents, AI companions, AI creative partners, AI reasoning systems. These are not faster horses. They are genuinely new categories. And the strategic challenge is structurally identical to Apple's in 2007: expressed consumer demand does not exist, so demand capture is not a viable path. The only viable strategy is demand creation.
Yet the dominant brand strategy for AI companies is still features first.
OpenAI's ChatGPT launch in November 2022 is instructive precisely because it worked without brand architecture. One million users in five days, a figure documented in early 2023 analyst assessments, is a demand-creation event by any measure. But it worked because the product itself was viscerally arresting. The first conversation with a capable large language model is a category-defining experience. You feel it in the moment.
Here is the problem. Visceral product experience is a fragile form of demand creation. When competitors match the capability, the demand that was built around the experience becomes available to whoever captures it. OpenAI's best demand-creation moment was the interface, not the positioning. The lack of durable brand architecture around that moment is precisely why dozens of nearly identical AI products can now make a credible claim to the same use cases.
The AI companies that will define their categories long-term are not the ones with the best benchmark scores right now. They are the ones currently building the trust reservoir that a future product will draw against. They are the ones constructing a worldview that does not mention the model architecture. They are the ones treating brand strategy not as a launch brief but as a decade-long architecture project.
The benchmark trap is just the demand capture trap in a new outfit. You are not reaching the audience that matters. You are speaking to the room that has already decided.
What I see AI founders doing right now, consistently, is spending their brand budget on the already-converted. Optimising launch messaging for developers and early adopters who were going to find the product anyway. Posting benchmark comparisons for an audience that already believes AI matters. That is not brand strategy. That is channel management mistaken for positioning. The audience that actually builds a category is the one that does not yet have a mental model for what you do. Nobody is talking to them.
Frequently Asked Questions
What is demand creation in brand strategy, and how is it different from demand generation?
Demand creation is the act of building desire for a product the market does not yet know it wants. It is a brand architecture strategy, not a marketing tactic. Demand generation, by contrast, accelerates and activates desire that already exists. Most marketing activity is demand generation. Demand creation is rarer, harder, and only possible when the brand architecture underneath it makes the leap credible.
Did Apple actually do market research before launching the iPhone?
Apple did conduct user research, but not to validate a known consumer need. Walter Isaacson documents that Jobs's approach was to study behaviour and friction, not stated preferences. Jobs held that consumers cannot articulate desires for things they have never imagined. The research question was never whether people wanted this. It was what was currently frustrating about the phones people already had.
Can brand strategy for AI companies actually create demand, or is product experience doing all the work?
Both matter, but product experience alone is fragile. ChatGPT's November 2022 launch created demand through visceral product experience, with no brand architecture underneath it. When competitors matched the capability, that demand became available to anyone. Durable demand creation requires brand architecture that makes the product experience culturally legible and personally meaningful, not just impressive in the moment.
What does brand architecture actually mean in the context of a tech startup?
Not a visual identity system. Brand architecture is the accumulated cultural meaning a brand has built over time: the trust reservoir, the worldview associations, the communities of belief that form around the company's position. For a startup, building brand architecture means making deliberate early investments in cultural presence and point of view, before the product conversation begins.
Build the Altar Before You Bring the Offering
On January 9, 2007, Jobs did not manufacture desire from nothing.
He activated desire that had been under construction for ten years: through the iMac, through the iPod, through iTunes, through every keynote that had trained an audience to believe that when Jobs called something revolutionary, it probably was. The iPhone was the offering. The altar had been built since 1997.
The practical implication is uncomfortable for anyone working on a product timeline.
For any brand attempting demand creation, and particularly for brand strategy for AI companies operating in categories that do not yet have names, the question is not how do we launch. It is: what have we built that gives us the right to ask for this kind of trust?
If the answer is nothing, the launch problem is the wrong problem.
The brands that will define the AI era are not the ones with the fastest release cycles or the most impressive benchmark sheets. They are the ones quietly building the altar right now, product decision by product decision, cultural statement by cultural statement, year by year.
The product comes later. The trust comes first.
