Altman faces $1.4 trillion skepticism: As long as computing power is still in short supply, OpenAI must continue to burn cash

OpenAI CEO Sam Altman emphasized that as long as there is a shortage of computing power, OpenAI must continue to burn money.

On December 19, OpenAI CEO Sam Altman explained in an interview on the Big Technology Podcast that the company's current losses stem from an aggressive expansion model training scale. As revenue grows and inference accounts for an increasing proportion of the computing cluster, it will eventually cover the training costs.

Faced with a massive gap between a $1.4 trillion spending commitment and $20 billion in revenue, Altman acknowledged that the company's training cost growth still exceeds revenue growth. However, he emphasized that OpenAI is always in a "computing power deficit" state, which precisely proves strong demand.

Altman stated that external concerns are only reasonable when the company has a large amount of idle computing power that cannot be monetized profitably. Currently, the severe shortage of computing power significantly limits the company's revenue growth potential, which is the core reason for continuing large-scale investments.

The path to profitability for Altman depends on a simple bet: OpenAI must find buyers as quickly as it develops. Ultimately, this bet will either continue to win or exhaust all resources.

Training Costs Dragging Down Current Profits

In the interview, when asked when the company could achieve breakeven, Altman did not shy away from the fact that expenses are still expanding.

Altman explained that as revenue grows, the proportion of the inference business in the entire fleet will increase, ultimately consuming the training costs. This is OpenAI's plan: to spend heavily on training but earn more returns through inference.

Altman candidly stated:

If we did not continue to significantly increase training costs, we would have been profitable long ago. But what we are doing now is investing heavily in training these large models.

The current financial situation is based on a clear bet, that is, it is necessary to continue investing heavily in model training to maintain technological advantages and market expansion.

Data shows that OpenAI is reportedly facing about $120 billion in losses before achieving profitability in 2028 or 2029. In response, Altman confirmed the company's strategic focus: to use revenue growth to support computing power expansion, rather than shrink due to short-term losses.

Computing Power is the Lifeline

In the face of skepticism about the huge gap between $1.4 trillion in spending and $20 billion in revenue, Altman's initial response seemed somewhat confused.

Altman talked about how humans find it difficult to intuitively understand exponential growth, stating that "evolution has made us good at many mental calculations, but building a quick mental model for exponential growth seems not to be among them."

Altman described computing power as the "lifeline" that facilitates everything. He stated that the company believes it can maintain a steep revenue growth curve, and currently, all signs indicate that this goal cannot be achieved without sufficient computing power.

He emphasized that OpenAI is "extremely constrained by computing power," which directly affects revenue performance Altman admitted, "We have always been in a state of computing power deficit, which has consistently limited what we can do," and he expects this situation to persist, although he hopes it will ease over time.

Altman revealed that the company has planned for this from multiple angles and anticipates that as technology advances, the efficiency of computing power per dollar will improve