Coffee Break: OpenAI as The Money Pit
OpenAI has fatally flawed finances which reveal a company that presents no serious threat to any of the tech incumbents, including Google.
In the (excellent) comments to my post last Monday on Google’s inexplicable-to-me decision to risk their search monopoly by going all-in on LLM AI one from Hickory exposed that I had failed to make a key point: ChatGPT is not a threat to replace Google as the leader in search because OpenAI loses money on every ChatGPT prompt and is trying to make it up in volume.
Let’s put aside the claims about LLMs having reasoning ability for now and focus on the bullish business case for ChatGPT as a killer app that threatens Google search. Cal Newport lays out that case:
The application that has… leaped ahead to become the most exciting and popular use of these tools is smart search. If you have a question, instead of turning to Google you can query a new version of ChatGPT or Claude. These models can search the web to gather information, but unlike a traditional search engine, they can also process the information they find and summarize for you only what you care about. Want the information presented in a particular format, like a spreadsheet or a chart? A high-end model like GPT-4o can do this for you as well, saving even more extra steps.
Smart search has become the first killer app of the generative AI era because, like any good killer app, it takes an activity most people already do all the time — typing search queries into web sites — and provides a substantially, almost magically better experience. This feels similar to electronic spreadsheets conquering paper ledger books or email immediately replacing voice mail and fax. I would estimate that around 90% of the examples I see online right now from people exclaiming over the potential of AI are people conducting smart searches.
This behavioral shift is appearing in the data. A recent survey conducted by Future found that 27% of US-based respondents had used AI tools such as ChatGPT instead of a traditional search engine. From an economic perspective, this shift matters. Earlier this month, the stock price for Alphabet, the parent company for Google, fell after an Apple executive revealed that Google searches through the Safari web browser had decreased over the previous two months, likely due to the increased use of AI tools.
Keep in mind, web search is a massive business, with Google earning over $175 billion from search ads in 2023 alone. In my opinion, becoming the new Google Search is likely the best bet for a company like OpenAI to achieve profitability…
That’s a seemingly reasonable claim, but doesn’t hold up after a look into OpenAI’s business plans for ChatGPT.
Despite their seeming threat to Google search, OpenAI is the kind of self-defeating competition no monopolist should fear, much less destroy its proven business model to compete with.
The New York Times put it pretty well last September:
— Nat Wilson Turner (@natwilsonturner) June 30, 2025
Morningstar summed up the NYT’s reporting well:
Financial documents reviewed by The New York Times reveal a company burning through cash at an alarming rate, raising questions about the sustainability of its current trajectory and the potential risks of prioritizing break-neck expansion over responsible AI development. Let’s discuss some of the key points from the New York Times report, which was published last week before the funding announcement:
— OpenAI’s monthly revenue hit $300 million in August 2024, a 1,700% increase since early 2023.
— The company expects to generate around $3.7 billion in annual sales this year and anticipates revenue ballooning to $11.6 billion in 2025.
— Despite rising revenues, OpenAI predicts a loss of about $5 billion. this year due to high operational costs, biggest of which is the cost of computing power it gets through its partnership with Microsoft.
— OpenAI predicts its revenue will hit $100 billion in 2029.
The Times report raises serious questions about OpenAI’s sustainability and realistic goals. The company’s monthly revenue growth from early 2023 to August 2024 is nothing short of explosive; however, the long-term projection of $100 billion in revenue by 2029 appears unrealistic. This figure would require sustaining an average annual growth rate of more than 90% for five consecutive years (93.3% to be precise, from an expected $3.7 billion in 2024 to $100 billion in 2029), a feat rarely achieved in the tech industry, especially for a company already operating at such a large scale. While impressive on paper, said projections may be masking underlying financial challenges and setting expectations that could be difficult, if not impossible, to meet.
Financial challenges become even more apparent given the current expense structure in relation to projected growth. It’s crucial to note that, even if it reaches the projected revenue targets, OpenAI is not merely failing to break even in 2024 – it’s losing significantly more money than it’s generating. This means that before OpenAI can even consider achieving its ambitious growth targets, it must first find a way to become profitable, or at the very least, break even.
Bryan McMahon pointed out the massive financial risk posed by the stock market bubble driven by faith in LLMs or as he calls it Generative AI:
Venture capital (VC) funds, drunk on a decade of “growth at all costs,” have poured about $200 billion into generative AI. Making matters worse, the stock market’s bull run is deeply dependent on the growth of the Big Tech companies fueling the AI bubble. In 2023, 71 percent of the total gains in the S&P 500 were attributable to the “Magnificent Seven”—Apple, Nvidia, Tesla, Alphabet, Meta, Amazon, and Microsoft—all of which are among the biggest spenders on AI. Just four—Microsoft, Alphabet, Amazon, and Meta—combined for $246 billion of capital expenditure in 2024 to support the AI build-out. Goldman Sachs expects Big Tech to spend over $1 trillion on chips and data centers to power AI over the next five years. Yet OpenAI, the current market leader, expects to lose $5 billion this year, and its annual losses to swell to $11 billion by 2026. If the AI bubble bursts, it not only threatens to wipe out VC firms in the Valley but also blow a gaping hole in the public markets and cause an economy-wide meltdown.
But wait it gets worse, per Ed Zitron:
It seems, from even a cursory glance, that OpenAI’s costs are increasing dramatically. The Information reported earlier in the year that OpenAI projects to spend $13 billion on compute with Microsoft alone in 2025, nearly tripling what it spent in total on compute in 2024 ($5 billion).
This suggests that OpenAI’s costs are skyrocketing, and that was before the launch of its new image generator which led to multiple complaints from Altman about a lack of available GPUs, leading to OpenAI’s CEO saying to expect “stuff to break” and delays in new products. Nevertheless, even if we assume OpenAI factored in the compute increases into its projections, it still expects to pay Microsoft $13 billion for compute this year.
This number, however, doesn’t include the $12.9 billion five-year-long compute deal signed with CoreWeave, a deal that was a result of Microsoft declining to pick up the option to buy said compute itself. Payments for this deal, according to The Information, start in October 2025, and assuming that it’s evenly paid (the terms of these contracts are generally secret, even in the case of public companies), this would still amount to roughly $2.38 billion a year.
I’ll let the Entertainment Strategy Guy nail the profitability coffin shut:
By all accounts, right now, OpenAI is losing money. Like literally billions of dollars. The energy costs of LLMs are enormous. If they’re pricing their services below market value, trying to gain market share, then we don’t know if AI can make money for the service it’s providing right now.
Two factors are driving these costs. First, the memory an AI program uses (either the more data it stores as it thinks or the longer it thinks about a problem/answer), the more it costs the AI companies in compute. Second, the AI companies are racing to build next-generation models that will require even more training, which means higher costs. And the salaries for top AI engineers/scientists are also sky-rocketing up.
This is why I’m somewhat skeptical about the sorts of things that OpenAI is promising that AI can do (like become your universal assistant that remembers everything about you); it seems like an absolute memory boondoggle of monumental proportions. How much energy will it take for AI to analyze my whole life if it’s already too taxing for an LLM to remember how to format links properly?
But wait, there’s even more bad news that just dropped. OpenAI is now in a bidding war for talent with some of the stupidest money out there, Mark Zuckerberg of Meta:
…competition for top AI researchers is heating up in Silicon Valley. Zuckerberg has been particularly aggressive in his approach, offering $100 million signing bonuses to some OpenAI staffers, according to comments Altman made on a podcast with his brother, Jack Altman. Multiple sources at OpenAI with direct knowledge of the offers confirmed the number. The Meta CEO has also been personally reaching out to potential recruits, according to the Wall Street Journal. “Over the past month, Meta has been aggressively building out their new AI effort, and has repeatedly (and mostly unsuccessfully) tried to recruit some of our strongest talent with comp-focused packages,” Chen wrote on Slack.
And speaking of dumb money and OpenAI, Softbank is involved, although maybe not as much as reported:
(In April) OpenAI closed “the largest private tech funding round in history,” where it “raised” an astonishing “$40 billion,” and the reason that I’ve put quotation marks around it is that OpenAI has only raised $10 billion of the $40 billion, with the rest arriving by “the end of the year.”
The remaining $30 billion — $20 billion of which will (allegedly) be provided by SoftBank — is partially contingent on OpenAI’s conversion from a non-profit to a for-profit by the end of 2025, and if it fails, SoftBank will only give OpenAI a further $20 billion. The round also valued OpenAI at $300 billion.
And things might not be going so well with Softbank because OpenAI is now talking to even dumber, and much more dangerous, money: Saudi Arabia.
And if you’ve ever paid attention to the actual words coming out of OpenAI CEO Sam Altman’s mouth, you’ll realize Altman attracting dumb money is just a case of birds of a feather flocking together.
Ed Zitron chronicles some of the stupid in his latest newsletter:
Here is but one of the trenchant insights from Sam Altman in his agonizing 37-minute-long podcast conversation with his brother Jack Altman from last week:
“I think there will be incredible other products. There will be crazy new social experiences. There will be, like, Google Docs style AI workflows that are just way more productive. You’ll start to see, you’ll have these virtual employees, but the thing that I think will be most impactful on that five to ten year timeframe is AI will actually discover new science.”
When asked why he believes AI will “discover new science,” Altman says that “I think we’ve cracked reasoning in the models,” adding that “we’ve a long way to go,” and that he “think[s] we know what to do,” adding that OpenAI’s o3 model “is already pretty smart,” and that he’s heard people say “wow, this is like a good PHD.”
That’s the entire answer! It’s complete nonsense! Sam Altman, the CEO of OpenAI, a company allegedly worth $300 billion to venture capitalists and SoftBank, kind of sounds like a huge idiot!
Ed also roasts Alphabet/Google’s Sundar Pichai:
Sundar Pichai, when asked one of Nilay Patel’s patented 100-word-plus-questions about Jony Ive and Sam Altman’s new (and likely heavily delayed) hardware startup:
I think AI is going to be bigger than the internet. There are going to be companies, products, and categories created that we aren’t aware of today. I think the future looks exciting. I think there’s a lot of opportunity to innovate around hardware form factors at this moment with this platform shift. I’m looking forward to seeing what they do. We are going to be doing a lot as well. I think it’s an exciting time to be a consumer, it’s an exciting time to be a developer. I’m looking forward to it.
The fuck are you on about, Sundar? Your answer to a question about whether you anticipate more competition is to say “yeah I think people are gonna make shit we haven’t come up with and uhh, hardware, can’t wait!”
While I think Pichai is likely a little smarter than Altman, in the same way that Satya Nadella is a little smarter than Pichai, and in the same way that a golden retriever is smarter than a chihuahua. That said, none of these men are superintelligences, nor, when pressed, do they ever seem to have any actual answers.
If ChatGPT were such an existential threat to Google’s search monopoly that Alphabet’s only option was risking the empire to beat OpenAI in the LLM race, it would be profitable or at least have a plausible path to profitability.
Sam Altman being a blithering idiot isn’t really the disadvantage it should be since he’s going up against competition like Mark Zuckerberg, Elon Musk, and Sundar Pichai.
This isn’t like Uber vs. the local taxi incumbents in the 2010s where despite Uber’s never going to be profitable business model they were able to take over in many markets because OpenAI does not have a huge cash advantage over Alphabet and never will.
Next week we’ll look at at Meta, the absolute stupidest tech money around — a company that put Dana White of the Ultimate Fighting Championship on its board.
And because I promised I’d get around to this, regarding LLMs ability to “reason” it was thoroughly debunked last October by six researchers from Apple in a paper called “GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models.”
When the paper was released the senior author, Mehrdad Farajtabar, tweeted that, ““we found no evidence of formal reasoning in language models …. Their behavior is better explained by sophisticated pattern matching—so fragile, in fact, that changing names can alter results by ~10%!”
