OpenAI Missed Revenue and User Targets as Competition Gains Ground

OpenAI, the company that ignited the generative AI boom, reportedly missed its internal benchmarks on both revenue and user growth at the end of 2025. According to Motley Fool, chief financial officer Sarah Friar has expressed concerns about the company's ability to afford its current contracts for computing resources if revenue doesn't pick up.

The missed targets come at a pivotal moment. OpenAI is planning to spend hundreds of billions of dollars on computing capacity to train future models and serve its growing user base. A revenue shortfall puts that spending plan under a microscope — particularly the compute contracts already signed with cloud providers.

The report weighed heavily on semiconductor stocks and OpenAI's biggest cloud infrastructure partners, signaling that Wall Street is paying attention to the gap between AI ambition and financial reality.

One reason for the shortfall is clear: competition is gaining momentum. Anthropic has reached 30.6% of U.S. companies paying for its services, according to a Ramp survey cited by Motley Fool, rapidly closing the gap with OpenAI's 35.2% — a level that has been stable for the past year.

Anthropic's enterprise‑focused strategy is paying off. After releasing Claude Code last year for software development, it built on that foundation with Claude Cowork for general knowledge work and developed specialized solutions for cybersecurity, finance, and legal sectors. The result is a product lineup that competes directly with OpenAI's ChatGPT across multiple verticals, not just coding.

On the other flank, Google's release of Gemini 3 last fall was seen as a major competitive threat. CEO Sam Altman reportedly issued a "Code Red" memo after Gemini outperformed OpenAI's leading model in benchmark tests. Apple chose to partner with Google for its Siri revamp, abandoning its own AI model in the process — a loss of a potentially massive distribution channel for OpenAI.

The timing of the revenue miss report coincides with a major shift in OpenAI's relationship with Microsoft. The two companies have loosened ties over the years, and the most recent amendment to their agreement removes the exclusivity of OpenAI products on Microsoft's Azure cloud platform. In exchange, Microsoft no longer pays a revenue share to OpenAI, per Motley Fool.

This opened the door for OpenAI to partner with Amazon and Google beyond infrastructure deals. The latest versions of GPT are now available through Amazon Bedrock, and OpenAI's coding agent Codex is also listed. These new channels could generate revenue that offsets the loss of the Microsoft revenue share.

But the shift also means OpenAI is now competing more directly for cloud partnerships — and its missed user targets suggest the brand alone may not be enough. For builders, the takeaway is that the AI model market is fragmenting, and betting on a single provider is riskier than it was a year ago.

OpenAI's situation illustrates a structural problem facing every frontier AI company. Training and serving large language models requires massive and growing compute commitments. Those commitments are signed years in advance, based on growth projections. If the projections don't materialize, the company is left with expensive infrastructure contracts and insufficient revenue to cover them.

CFO Sarah Friar's reported concerns about affording compute contracts highlight the tension. OpenAI has committed to spending hundreds of billions on data center capacity. If user growth stalls while Anthropic and Google continue gaining share, those commitments become a liability rather than an asset.

Meanwhile, Amazon and Alphabet — both Anthropic investors — could benefit regardless of which AI company wins. As TNW reported, both cloud giants are booking massive paper gains on their Anthropic stakes while AWS and Google Cloud capture the compute revenue from Anthropic's growth. OpenAI's struggles may simply accelerate the flow of enterprise AI spending toward Anthropic's infrastructure providers.

The OpenAI revenue miss isn't an existential crisis — the company remains the AI market leader with 35.2% enterprise penetration. But it signals that the AI market is maturing beyond a single‑vendor story. Builders who anchored their workflows to OpenAI should pay attention to three shifts.

First, Anthropic's enterprise traction is real and accelerating. The 30.6% figure from Ramp's survey suggests Claude is becoming a default option for businesses — not just a developer curiosity. Second, Google's Gemini 3 is competitive on benchmarks and winning platform deals like Apple's Siri integration, giving it distribution that OpenAI lacks. Third, the Microsoft-OpenAI uncoupling means the Azure exclusivity advantage is gone — OpenAI now competes on model quality alone.

For builders evaluating which AI platform to build on, the old assumption that the assumption that OpenAI is the safe default deserves a second look. The market is fragmenting, and the company that defined the category is showing signs of being catchable.