Varying accounting treatments for cloud-platform revenue — applied differently by OpenAI and Anthropic — have left side-by-side financial comparisons of the two firms on uncertain ground. Revenue figures reported by Anthropic draw on transactions processed through partner platforms, among them Amazon Web Services and Google Cloud, while revenue accounting at OpenAI draws a distinction when tabulating certain sales generated through partner arrangements.

For every dollar flowing in through those distribution channels, roughly sixteen cents goes back to Anthropic's cloud partners as their share of the arrangement. A Reuters analysis found that those transactions generated roughly half of the company's total revenue in the preceding year.

Before any side-by-side analysis of the two firms can carry real weight, their figures need to be placed on the same accounting basis — a step that becomes unavoidable once the gap between their respective reporting practices is taken into account. No comment from OpenAI reached the outlet before publication.

Investor disclosures formed the basis of a revised estimate that one prominent financial outlet subsequently brought to public attention. Heading into 2026, OpenAI's annualised revenue run rate had climbed to USD 20 billion, more than three times the USD 6 billion pace recorded two years earlier.

Heading into the third month of that quarter, Anthropic had sharpened its competitive position enough to press the company on multiple fronts. Quarterly figures placed OpenAI's revenue at USD 6.7 billion for the period, while Anthropic's tally over the same stretch reached USD 11.5 billion.

Anthropic's annualised revenue run rate passed the USD 65 billion mark during July, while unnamed sources had told a news outlet that the figure might climb to USD 100 billion before 2026 draws to a close. Rivalry among providers targeting corporate clients has grown more pronounced as generative AI tools find wider adoption across the business world.

Revenue growth ambitions run alongside the pressure of steep outlays on computing infrastructure and AI model development at both firms. Both OpenAI and Anthropic are moving toward potential stock market listings around the time this updated valuation figure surfaced. Public disclosure through a listing would allow investors to examine how the company accounts for revenue, what it spends to operate, and whether its finances can hold up over time.

Among technology companies navigating rapid expansion, the annualised revenue run rate has held its place as a widely used performance benchmark. A single month's worth of sales data gets scaled up to a full-year projection by multiplying that figure across twelve months. That metric captures only a partial revenue picture, and when monthly sales vary, the figure it produces can give investors a distorted read on the business's full annual performance.