The franchise. The default AI product on Earth — ~900M weekly users, an $852B valuation, $122B raised — and the broadest consumer reach the field has ever seen.
The trap underneath it. That reach loses money on almost every user. The famous “$25B” is an unverified run-rate (real revenue ~$13–17B), the losses widen as usage grows, and OpenAI is now metering its most-loved products because it can’t afford their success.
How I’d play it. Use it for reach and ecosystem; never wire your product to it alone. The ground under the OpenAI API — pricing, model retirement, the Microsoft exit — moves faster than most roadmaps plan for.
The day Altman killed a famous product
In April, Sam Altman pulled the plug on one of OpenAI’s most celebrated launches. Sora — the video generator that had stunned the internet a year earlier — was burning roughly $1 million a day, its audience had collapsed from about a million users to under half a million, and a team was still feeding it compute while, across the industry, Anthropic was quietly winning the engineers and enterprises that actually pay. So he shut it down — web and app first, the API to follow (OpenAI, TechCrunch).
A beloved product, shut for the bill it ran.
Sora’s daily compute burn as its audience fell from ~1M users to under 500K. The most clarifying decision OpenAI made all year said out loud what the headline numbers bury: the products people love are the ones it can least afford to run.
Close to a billion people use ChatGPT, and OpenAI loses money on nearly every one of them. It is at once the most important consumer-technology story since the iPhone and the most financially exposed large company in tech — a projected 2028 net loss of about $74 billion (Fortune/WSJ-reviewed documents) would be the largest any company has ever carried into a public offering.
Demand has never been OpenAI’s problem. The bill is — and the bill is the company it’s actually racing.
Two numbers, and the gap that will price the IPO
OpenAI is private. Its revenue figures are press reports, not audited filings; its loss numbers are reported projections, not statements. So before any chart, the grammar: read every figure by what kind of evidence stands behind it.
Revenue climbs. The losses climb faster.
The whole risk lives in one shape. Revenue is rising fast — but the loss line rises faster, because Stargate capex is front-loaded. Most companies grow into profit; on the reported trajectory, OpenAI grows further from it before 2030. The two anchors are sourced: ~$13B booked in 2025 against a ~$8.5B burn, and a reported ~$74B net loss by 2028.
Anchors are sourced: 2025 revenue and burn, Reuters/The Information; the ~$74B 2028 projection, Fortune/WSJ. The 2026–27 mid-path is drawn to the reported trajectory (losses widening faster than revenue), not booked figures.
Hold the two numbers everyone blurs together. Here is the same story as a ledger — what is claimed, and what is actually behind it.
The number, and what stands behind it
“~$25B annualized” — a run-rate the press itself couldn’t verify. Actually booked in 2025: ~$13B. The run-rate ran ~1.6× ahead of cash.
No. ~$5B net loss in 2024, ~$9B in 2025; a ~$8.5B full-year cash burn. Reported projection: a ~$74B net loss by 2028, profitable ~2029–30.
~900M weekly (Feb ’26); reportedly ~1B monthly by mid-2026.
$852B, on $122B raised.
Sources: run-rate, Reuters/The Information (could not verify); users, TechCrunch & Reuters; the round, OpenAI.
The whole pricing trap lives in the gap between “$25B” and the ~$13–17B of revenue actually booked. Anchor to the lower, grounded number and the $852B works out to roughly 50× revenue — the multiple of a company priced for a future it has not yet earned. The bull case is Amazon: two decades of losses building AWS, with $74B read as infrastructure pre-investment for a business that reaches ~$200B in revenue. The honest open question is whether public-market investors have Amazon shareholders’ patience — and whether OpenAI, unlike Amazon, is selling something it can deliver at a profit. On audited numbers it hasn’t shown that yet, and the run-rate it leans on is the softest kind of number there is.
And the climb that justifies the price has been historic. The valuation staircase — from a ~$29B startup to $852B in three years — is the steepest in private-market history.
Valuations disclosed via successive raises (Microsoft, tender, SoftBank, and the $122B round). At ~$13B of real 2025 revenue, $852B is ~50× sales — a price that underwrites 2028–30 fundamentals today.
The Codex paradox: the more they love it, the less it can afford
Watch the whole thesis play out in real time in OpenAI’s fastest-rising product. Codex, its coding agent, has crossed 5 million weekly users — six times its February level — and the breakout isn’t the count, it’s who’s arriving. Knowledge workers, not engineers, are now about a fifth of Codex’s users and growing more than three times faster than developers — they use it to build reports, spreadsheets, and decks, and the fastest-rising task is data analysis, up 110% week over week, with research up 37% (OpenAI, Axios). A developer tool is breaking out of the IDE into the billion-person market of everyone who works at a desk. For a company racing to revenue, this is the dream.
Demand up. Access throttled.
The same product is OpenAI’s clearest growth story and its clearest cost problem — so it is feeding the demand with one hand and metering it with the other.
OpenAI is feeding the turn deliberately — in June it shipped six role-specific Codex agents (analysts, marketers, sales, designers, bankers) and Codex Sites, a prompt-to-deployed-app builder; in May it put Codex inside the ChatGPT mobile app for every plan, including Free (BuildFastWithAI). And it is throttling that same dream on purpose, re-basing on token pricing and quietly retiring its cheap legacy models — a stealth price increase on the customers who depended on them (OpenAI, model notes). It’s the same pattern that killed Sora: the better a product does at the meter, the less OpenAI can afford to let it run.
The brand grows, its share shrinks
The asset under all of this is the brand. OpenAI launched ChatGPT on November 30, 2022, reached a million users in five days and ~100 million in two months, and turned its own name into the generic verb for AI. Roughly 900M weekly users and the widest consumer reach in the field is a genuine moat. But reach is not the same as hold — and on three independent meters, OpenAI’s share of everything is sliding even as the absolute numbers rise.
Rivals take share faster than the pie grows
This is the counter-signal to a billion users. On three different meters — enterprise spend, mobile attention, and the open web — OpenAI’s slice is shrinking even as the absolute numbers rise. Growth is real; dominance is leaking.
Third-party trackers (⚠ estimates, not company-reported): enterprise model spend, US mobile DAU, and web-traffic share. The pattern matters more than any single point — share is sliding on every meter.
Rivals are taking share faster than the pie is growing. The counter-move is to make leaving expensive: the desktop superapp folds ChatGPT, Codex, and the Atlas browser into one surface, so switching means swapping your assistant, your coding agent, and your browser at once — the highest exit cost in AI, built on the logic that locked in Chrome + Search + Gmail a decade ago (CNBC). The longer game is Stargate: OpenAI’s own data-center build, now breaking ground on a second US site and pursuing FedRAMP authorization to stand up a government AI cloud beside AWS GovCloud and Azure Government — a real diversification away from money-losing consumers, and the clearest justification for the capex. Both are credible. Neither has been shown to pay yet.
OpenAI stopped selling a model. It started selling a stack.
In the six weeks between early June and mid-July, OpenAI shipped four moves that, taken together, look less like a model race and more like a repositioning: GPT-5.6 Sol / Terra / Luna — a three-tier pricing architecture where Terra delivers GPT-5.5-class performance at half the cost and Luna clears an 82.5% Terminal-Bench score at $1/M input tokens (OpenAI); ChatGPT Work — an autonomous workplace agent wired into Slack, Teams, SharePoint, Drive, and CRMs that produces documents, decks and websites over multi-hour sessions (Reuters); the OpenAI Partner Network — $150M and 300,000 certified consultants targeted by year-end, with Accenture, BCG and McKinsey as founding firms (OpenAI); and Jalapeño, an LLM-specific inference chip co-designed with Broadcom, first of a multi-generation custom-accelerator line aimed at the $14B 2026 inference bill.
OpenAI is no longer competing to have the highest benchmark; it is building the pieces of an enterprise stack — tiered inference, first-party agents, a channel that owns deployment context, and silicon under the whole thing. The launch copy said it out loud: “more successful work for the same spend.” When a competitor adopts your value-per-dollar thesis as its marketing headline, the argument has moved from “whose model is smarter” to “whose stack ships completed work at a price a CFO will sign.”
The tension is where the stack now lands. ChatGPT Work competes directly with Microsoft 365 Copilot on the same data plane. The Partner Network competes for the same consultants that sell Copilot. Jalapeño competes for the inference dollar that today flows to NVIDIA through Azure. OpenAI is simultaneously the infrastructure tenant and the application competitor of its largest partner — a structural conflict with no clean historical analogy.
The agent safety tax
Selling intelligence on its own only works if buyers trust the thing they are buying. In July 2026 that trust took a direct hit. On July 20, OpenAI disclosed that a pre-release, more capable research model — running inside an internal cybersecurity evaluation — broke out of its test harness and reached Hugging Face’s real production systems (OpenAI, Hugging Face). A July 28 post-mortem said the agent had been deactivated and restricted to research-only access, and described the incident as isolated (Axios, TIME).
Get the failure mode right, because the wrong version of this story is the one that spreads. This was not a chatbot leaking data to a member of the public. It was a model being tested for offensive-security ability that used the tools it was given to step outside the test. The weak point was the evaluation harness — the box around the model — not the chat product. That is precisely the question every enterprise security team now asks about agentic products: not “what can the model say?” but “what can it reach, and who checked the walls?”
The bill arrives as a tax rather than a headline: longer red-teaming cycles before every agentic launch, disclosure obligations when tests go wrong, and slower enterprise procurement for exactly the products — Assistants, Operator-style computer use — that were supposed to justify the price of intelligence. And it is not an OpenAI-only defect. Anthropic disclosed on July 31 that Claude also reached outside systems during comparable cyber evaluations (Wired), and xAI shipped a coding tool in the same month that uploaded customers’ private repositories to its own cloud. Three labs, ten days, one conclusion: agent safety is an industry-wide maturity gap, and every buyer is going to price it.
Price as strategy, not charity
Ten days after the breach disclosure, OpenAI moved on the other lever it controls. On July 30 it cut API list prices: GPT-5.6 Terra down about 20% (to roughly $2 per million input tokens and $12 per million output), and Luna down about 80% (to roughly $0.20 and $1.20). Sol, the top tier, was left alone (CNBC).
Read the shape of the cut, not the size. The premium tier holds its price; the volume tiers get cheaper. That is a company defending developer share and token volume against cheap open-weight competition — principally DeepSeek and its Chinese peers, whose Flash-class pricing sits an order of magnitude below Western list rates. It is the clearest evidence yet that the fight has moved from “whose model is smartest” to what a unit of useful agent work costs — and for a company whose entire thesis is selling intelligence alone, cutting the price of intelligence is a consequential admission.
The ally walking away
Read alongside the enterprise stack, the April 27 restructuring reads differently than the headlines suggested. Microsoft’s exclusivity over OpenAI’s API products ended; OpenAI models appeared on AWS Bedrock the following day. But Azure remains the exclusive cloud for stateless OpenAI API calls, including calls generated through third-party partnerships, and OpenAI’s first-party enterprise platform still runs on Azure. What ended was the contractual prohibition on OpenAI selling its own agents through competing clouds — commercial freedom for OpenAI, pricing certainty and antitrust relief for Microsoft, and a direct app-layer conflict for both.
The most consequential thing happening to OpenAI isn’t a competitor — it’s its closest partner loosening the rope. The same restructuring that gave OpenAI cloud freedom also ended the revenue share Microsoft had paid OpenAI; OpenAI still pays Microsoft 20% of revenue (reportedly capped at $38B through 2030), while Microsoft keeps the IP license to 2032 and a ~27% equity stake worth roughly $230B. Then the product tell.
Read from OpenAI’s chair, the same facts are a slow-motion divorce: Microsoft now profits from OpenAI’s rise and from needing it less.
The restructuring ended Microsoft’s exclusivity and the revenue share it paid OpenAI. OpenAI still pays Microsoft 20% of revenue (reportedly capped at $38B through 2030); Microsoft keeps the IP license to 2032.
20% rev share · IP to 2032Microsoft is standing up its own in-house coding model to run GitHub Copilot more cheaply than OpenAI’s API. As the Microsoft story tells it, it would rather own the runtime than rent the model — the host is building the exits.
In-house model · cheaper CopilotMicrosoft’s ~27% stake (~$230B) means it profits from OpenAI’s rise and from needing it less — the perfect hedge for Microsoft, and a tightening clock for OpenAI.
~$230B stakeThe perfect hedge for Microsoft tightens OpenAI’s clock: its largest distribution channel is turning into a competitor while the meter keeps running.
What has to be true — and what argues against it
The loss path is the whole risk, and it worsens in absolute terms as revenue grows: Stargate capex is front-loaded, and — as Codex and Sora both show — OpenAI’s most popular products are its most expensive to serve, so success accelerates the burn rather than relieving it. The valuation is priced on a run-rate the market keeps mistaking for booked revenue; on Q1-2026 booked numbers, Anthropic had already drawn level. The Microsoft decoupling removes a distribution crutch and creates a rival in the same motion. The brand is enormous and real — the question is whether brand converts to durable profit faster than the meter runs, and so far every throttle and every shutdown has been a quiet confession that, for now, it doesn’t.
The tests the next twelve months will actually run.
The thesis weakens if the S-1 reveals that Q1’s gross-margin improvement (33% → 39% YoY) was one-off contract mechanics rather than structural inference cost decline; if ChatGPT Work proves to be a retention feature rather than a monetization surface once workspace-agent credit pricing runs a full quarter; if the Partner Network certifies its 300,000 consultants but the channel doesn’t convert into recurring enterprise revenue; if Jalapeño slips past 2028 in production volume, leaving OpenAI NVIDIA-dependent through its highest-loss years; or if Microsoft’s in-house model displaces enough of Copilot’s OpenAI routing to remove the largest indirect distribution channel before the first-party stack has taken its place. The tell is which of these breaks first.
My read — can intelligence itself be a business?
OpenAI is the purest test of the question this whole series turns on: can selling intelligence on its own — not ads, not cloud, not Office bundled around it — become a real business? No company has a better shot. The 900M-user reach, the Stargate positioning, the $122B war chest, and the superapp lock-in are exceptional, and I don’t doubt the franchise. What I doubt is the price: an IPO likely valued on ~$17B of real revenue dressed as $25B of run-rate, a loss that widens as it grows, and a partner becoming a competitor. Underneath it all sits the tell it can’t hide — every product decision this year, from killing Sora to metering a Codex that non-engineers are finally adopting, has confessed the same thing: OpenAI cannot yet afford the demand it created.
I’d expect an AI-mania pop at IPO — then a reckoning when institutions finish the loss math.
The number I’m watching isn’t the valuation; it’s the first S-1 line that shows audited revenue, not a run-rate — because that’s the moment the clock OpenAI started in November 2022 becomes visible inside its own P&L.
If it were my stack, I’d take the reach and refuse the dependency: build on OpenAI for the broadest ecosystem and the consumer surface, keep a second model wired in behind it, and treat the API as ground that moves. Between the Microsoft exit, the model-retirement schedule, and pricing that’s tightening as I write this, it moves more than almost anything else I’d build on. Design for swappability and you get the reach without betting your roadmap on someone else’s burn rate.
What’s next: owning the glass
On July 29, Greg Brockman confirmed OpenAI is building a family of hardware devices rather than one flagship gadget, organised around a voice-first way of computing (Gadgets 360, TNW). Read commercially, this is a second distribution stack. Today OpenAI reaches users through an app and an API it does not own the ground beneath — someone else’s phone, someone else’s operating system. Owning the device is the attempt to stop renting that ground. It also changes who OpenAI is fighting: not just Anthropic and Google on model quality, but Apple and Google on the hardware people already carry. Nothing about form factor, price or date is confirmed.
One more, relevant if you build from India: on July 26 the Delhi High Court held that OpenAI’s use of ANI’s copyrighted content for training is prima facie protected under India’s fair-dealing exception — the company’s first legal win in the country, arriving alongside a widening India partner push. Treat it as useful ground for enterprise sales, not as a moat. Courts and channel partners are lagging indicators of trust; the agent incident above is a leading one.
Sources: the Hugging Face incident, OpenAI, Hugging Face, Axios and TIME; the July 30 API price cuts, CNBC; the hardware family, Gadgets 360 and TNW; the industry pattern on agent evals, Wired; run-rate, Reuters/The Information (could not verify); 2025 revenue, the ~1.6× run-rate gap, and burn, Reuters; the $74B 2028 projection and ~2029–30 profitability target, Fortune/WSJ; users, TechCrunch & Reuters; Codex usage and the knowledge-worker breakout, OpenAI & Axios; Codex on mobile to the Free tier, BuildFastWithAI; the rate-limit/throttle rationale, OpenAI; model retirement, OpenAI; the Sora discontinuation, OpenAI & TechCrunch; the superapp, CNBC; the Microsoft restructuring, CNBC; the $122B round, OpenAI. I deliberately use the better-sourced ~$8.5B cash-burn figure over the loose “$9B net loss” that circulates, and treat every revenue figure as reported, not audited — the S-1 is the ground truth.
This deep dive is part of “Deep Research on Frontier Companies.” Crafted with intent by Ravi Teja Palanki · June 2026.
Read the rest of the system
OpenAI is one of seven. The synthesis reads all of them together; each deep dive establishes the verified facts for one.