This article first appeared on GuruFocus.
According to a report from The Information, OpenAI has been extracting meaningfully higher margins from its paid products this year as it works to defend its leadership in artificial intelligence. People familiar with the figures told the publication that OpenAI’s internal compute margin, which measures revenue left after model-running costs for paying users, reached 70% as of October, up from 52% at the end of 2024 and roughly double the level seen in January 2024. An OpenAI spokesperson said the company does not disclose these figures and declined further comment, but the improvement could suggest that recent monetization efforts across corporate and consumer offerings are beginning to gain traction, even as the company remains unprofitable.
At the same time, OpenAI continues to face pressure around spending discipline and intensifying competition. After Alphabet Inc.’s Google (NASDAQ:GOOG) Gemini model performed better on certain benchmarks, Chief Executive Officer Sam Altman reportedly called a code red, redirecting internal resources toward improving ChatGPT and slowing progress on an advertising initiative. Most users still rely on ChatGPT’s free tier, but OpenAI has been pushing its business-focused products and paid software features for industries such as financial services and education, where it competes directly with Google and privately held rival Anthropic.
The Information also reported that OpenAI’s paid accounts generate better compute margins than Anthropic’s, while Anthropic is said to operate with stronger overall server efficiency. Against that backdrop, OpenAI is in early discussions to raise at least $10 billion from Amazon.com Inc. (NASDAQ:AMZN) and potentially use its chips, a move that could value the company at north of $500 billion. These talks, alongside the reported margin gains, suggest OpenAI is attempting to balance revenue optimization with the heavy infrastructure demands required to remain competitive at the frontier of AI development.