Opelai has committed more than $1.4 trillion to build its data center infrastructure over the next 8 years. It is projected to lose $74 billion in 2028 alone, according to a new report from the Wall Street Journal citing internal documents from the company. At the opposite end of the spectrum, aistrup anthropic auppic meets part of the hype of sam Altman’s company enjoys, and will reportedly be able to break away that year.
The literature seems to suggest very different approaches to corporate profit. Anthropic, heavily funded by investors at a valuation of nearly $200 billion, has been quietly building a subscriber base for corporate customers. According to a report from the WSJ earlier this month, nearly 80% of Anthropic’s revenue comes from business customers, with more than 300,000 customers. OpenAI says that the subscribers are almost a million businesses and more than seven million chatgtrent work “seats” (a photo that could have been shared just days after receiving the anthropic numbers).
But, according to the magazine’s reporting, Opelai’s margins on subscriptions are more diluted than those of anthropic. One would think, given that, accai could spend its money without limits until it reaches a clear path to profitability. Instead, the company is going all out in building data centers and submerging chips to train and provide its models.
There is also a burning money that has a plan to just get people to thank in its ecosystem in the hope that they will eventually not be run over by money. Earlier this week, Forbes reported that SORA 2, the company’s video generation model opened millions of people who create copyright infringement by waiting, worth $ 5 billion.
It seems that the way to open OpenAi is something close to the zirp-era of big tech, when firms were happy to run deep into the red because they were getting money with zero interest. That led to an Uber-like approach to driving prices down to unbelievably low levels to capture as many audiences as possible and finally flip the switch on the city.
It’s not clear if the method actually exists for Opelai, but it seems to think it works. Per WSJ, the company’s internal documents show a rapidly growing capital base that will lead to the company turning a profit by 2030, which would be no more than an incident in its eight-year commitment to build expensive data operations. On the way to the black end of its bottom line, it is reported to burn about 14 times as much money as anthropic.
The problem for anthropic and any other company working in the AI space is that Opelai can make everyone take everyone with it. The company’s financial commitment is firmly tied to many key players in the AI field, and its failure can lead to major positive consequences. It’s no surprise that OpenAI is releasing government guarantees on its use.
