The artificial intelligence revolution has created a stark divide in the technology world. While established tech giants report soaring profits driven by AI-related revenue, the very startups pioneering generative AI are bleeding billions of dollars in losses.
The Profit and Loss Divide
Recent quarterly results reveal an unprecedented financial chasm. Nvidia, Alphabet, Amazon and Microsoft have seen their profits skyrocket as AI-related revenue pours in. Meanwhile, Meta's profits were largely wiped out by tax considerations. This impressive performance has provided comfort to investors concerned about AI valuations becoming overinflated.
However, this success story has what industry experts call an "ugly underbelly" - the massive, ongoing losses at generative AI startups that are spending enormous amounts on chips and data centers supplied by these profitable public companies.
Startup Losses Fuel Big Tech Gains
The financial dynamics are clear: much of the AI-related profits at established companies come from being suppliers to, or investors in, the private companies building large language models. These startups, including industry leaders OpenAI and Anthropic, are losing money as rapidly as they can raise it, with plans to continue this pattern for years.
OpenAI alone lost more than $12 billion in the quarter ending September 30, according to calculations based on Microsoft's share of the losses. This staggering amount matches the world's largest corporate loss reported so far this quarter - that of satellite communications company EchoStar. However, EchoStar's loss resulted from a $16.5 billion noncash charge, while OpenAI's represents real money being spent.
The scale becomes even more dramatic when considering that OpenAI's quarterly loss equates to 65% of the combined increase in underlying earnings (before interest, tax, depreciation and amortization) of Microsoft, Nvidia, Alphabet, Amazon and Meta.
The Road to Profitability: A Distant Dream
Current financial projections suggest the bleeding will continue for years. OpenAI doesn't expect to turn profitable until 2030, while Anthropic is targeting 2028. The company's own forecasts reveal:
- Revenue expected to grow from $13 billion this year to $30 billion next year
- Further doubling to approximately $60 billion by 2027
- Losses predicted to triple to more than $40 billion by 2027
- Financial balance not expected until computing costs level off in 2029
These forecasts come with significant uncertainty. OpenAI has only recently settled on its corporate structure, products remain in development, potential customers are still figuring out implementation strategies, and competition is evolving at breakneck speed.
The Spending Spree Continues
The financial commitments are extraordinary. OpenAI has committed to spend $250 billion more on Microsoft's cloud services and has signed additional massive deals:
- $300 billion agreement with Oracle
- $22 billion deal with CoreWeave
- $38 billion arrangement with Amazon
This insatiable demand for computing capacity explains why data center construction has shifted into overdrive globally, with significant implications for energy consumption and infrastructure development.
The Fundamental Challenge
For this financial model to sustain itself, two critical developments must occur. First, AI developers need to create winning products that generate sufficient revenue to cover their massive research and computing costs. Second, investors must continue financing these enormous losses until profitability is achieved.
The technology faces significant hurdles. While AI chatbots offer near-magical experiences, they still struggle with basic errors, security vulnerabilities, and the persistent problem of "hallucination" - making up facts. Solving these issues is crucial for broader business and individual adoption.
Industry veterans recall the old saying: "Revenue is vanity, profits are sanity, cash is reality." Currently, the AI sector is focused on the vanity of revenue growth, but investors may eventually shift their attention to what many consider insane losses.
The reality remains clear: the cash flowing from OpenAI and similar companies directly bolsters the earnings of big tech companies, creating an interdependent financial ecosystem that could prove vulnerable if investor enthusiasm wanes or fundraising becomes challenging due to economic downturns.
