Anthropic has moved ahead of OpenAI in annualized revenue in the first half of 2026. The Information reported on May 27 that Anthropic is nearing $450 billion in annualized revenue, while OpenAI stands at about $330 billion. On that basis, Anthropic is ahead by at least 35%.
The reversal happened quickly. According to the source material, Anthropic was generating only about $9 billion in annualized revenue at the end of 2025, less than half of OpenAI’s level at the time. In less than six months, Anthropic’s revenue increased about fivefold, while OpenAI grew by roughly 50%, turning a catch-up story into a leadership shift.
Enterprise demand and coding products drove Anthropic’s rise
The report links the change to a clear split in strategy. Anthropic’s growth came largely from enterprise AI, code generation, and white-collar workflow use cases. OpenAI, by contrast, remained more focused on consumer products, especially ChatGPT Plus subscriptions.
Enterprise buyers are paying for workflow automation rather than just faster language models. The source highlights several areas where Claude gained traction, including rising adoption of Claude Code and Claude Projects among enterprise developers, along with use cases ranging from document analysis to meeting notes. Businesses are also willing to pay a premium for predictable output, creating revenue with stronger stickiness and higher ARPU.
OpenAI faces heavier compute and product costs
OpenAI’s cost structure appears more demanding. The report says spending on compute and consumer operations kept rising, while ChatGPT subscription growth slowed. Customer acquisition costs in the consumer market also increased, and ARPU came under pressure from price competition.
There is also the burden of infrastructure investment. The source says OpenAI’s top-end compute spending, especially its dependence on Nvidia-related capacity, has not yet been fully reflected in revenue. At the same time, its lineup stretches from GPT-4o to Sora and the Agent ecosystem. That broader product range may expand reach, but the revenue contribution from each line does not necessarily scale at the same pace, leaving margins more diluted.
The revenue shift points to a broader change in AI demand
This change in ranking says more than who is first and second. It suggests the center of gravity in AI is moving away from consumer applications and toward enterprise productivity tools. As AI moves from chatbot use toward programming, analysis, and editing work, business customers are showing a higher willingness to pay than consumers.
The report also argues that Anthropic’s earlier push into enterprise markets, especially Claude’s reputation within developer communities, began paying off in 2026. OpenAI still has large revenue, but its wider consumer focus and longer product roadmap come with a more visible trade-off between growth and cost discipline.
Taiwan market shows uptake, but cost and deployment remain hurdles
The source also discusses enterprise AI adoption in Taiwan. Manufacturing, semiconductors, and 3C industries are in a phase where companies are trying to optimize capacity with AI, and enterprise tools are seeing higher penetration among developers and engineers. At the same time, smaller businesses remain sensitive to cost, while larger enterprises must weigh data privacy concerns against cloud deployment choices.
Compared with the US, Taiwan’s enterprise AI market is described as more focused on industry-specific applications than cross-industry platforms. The article also notes that government support for AI startups is helping speed up adoption. If manufacturing use cases move from pilot programs to scaled deployment, the revenue opportunity from enterprise AI could prove more durable.