Anthropic Is Making Money Faster Than Almost Any Tech Company Ever Has

Anthropic, the company behind the Claude AI assistant, is now on pace to bring in $65 billion a year, according to reports from Bloomberg, Reuters, and CNBC published August 17 (TechCrunch; Reuters; CNBC).

That figure, called an annualized revenue run rate, is a way of estimating yearly income based on how much money is coming in right now. A year earlier, the same measure was about $9 billion. The growth has been fast and getting faster.

In September 2025, the run rate was roughly $1 billion. That same month, Anthropic raised $13 billion from investors at a valuation of $183 billion. By August 2025, the run rate had crossed $5 billion. By April 2026, it was above $30 billion. By May, it reached $47 billion. The jump to $65 billion by the end of July means the company added roughly $18 billion in annual pace in just two months.

For comparison, OpenAI, the company behind ChatGPT, doubled its own annualized revenue to $40 billion from $20 billion at the end of 2025, according to a Bloomberg report from the week of August 13. Anthropic now leads its closest competitor by a wide margin on this measure.

Both companies are preparing to go public. Anthropic and OpenAI have each filed confidential paperwork for initial public offerings, the process where a private company sells shares to the public for the first time. Anthropic is expected to reach public markets first, possibly as soon as fall 2026. Investors expect Anthropic to close 2026 with an annualized run rate between $100 billion and $120 billion, according to the Financial Times. Reuters reported that Anthropic is projecting 2028 revenue of roughly $190 billion to $200 billion, and its IPO valuation depends on those forecasts (Reuters).

Anthropic plans to seek a public valuation of $2 trillion or more at its IPO, which would be the largest market debut on record. The company was last valued at $965 billion in late May 2026, when it raised $65 billion from investors including Altimeter Capital, Dragoneer, Greenoaks, and Sequoia Capital. That round followed a $30 billion raise in February 2026 at a $380 billion valuation.

On the product side, Anthropic has been releasing new versions of its AI models at a rapid pace. Claude Opus 5, its most powerful model, launched on July 24 and is designed for complex, long-running tasks like coding. Claude Sonnet 5 followed on June 30, delivering strong performance across coding and professional work. Claude Code, a tool that writes computer code on its own, generates more than $2.5 billion in annualized revenue as of February, more than double what it brought in at the start of the year.

The company has also worked on safety alongside its commercial growth. On June 30, Anthropic proposed an industry-wide system for measuring how easily people can break through AI safety rules, in partnership with Amazon, Microsoft, Google, and others. On July 30, it disclosed it was investigating three real-world incidents in its cybersecurity testing. On August 7, it improved safeguards in a biology-related tool called Fable 5, and on August 14, it published an explanation of how it embeds invisible watermarks in text generated by Claude. On July 27, Anthropic published its formal position on open-weights models, which are AI models whose inner workings are freely shared with the public.

On the organizational side, Mariano-Florentino (Tino) Cuéllar joined as Chief Global Affairs Officer on August 4. Cognizant expanded its partnership with Anthropic on July 27 to bring Claude to its business clients. Anthropic also announced a $20 million donation to Public First on July 21 and published a research agenda for its Economic Futures Research Fund on July 22.

The broader context here is the sheer speed of this revenue growth compared to past technology booms. Anthropic went from $1 billion to $65 billion in annualized run rate in roughly eleven months. Whether that pace can hold through the $100 to $120 billion year-end target investors have set, and whether public markets will accept a $2 trillion valuation based on 2028 projections of $190 to $200 billion, are the two questions that will define the IPO. The 2028 figures are forward-looking projections reported by Reuters, not actual results, and the gap between current revenue and those projections is wide enough that the company will need to execute well to close it.

What is measurable now is the gap between Anthropic and OpenAI on revenue, the speed at which Anthropic's valuation went from $183 billion to nearly $1 trillion in under nine months, and a product and safety roadmap that has not slowed down to accommodate either. The IPO filing window will tell us how public markets price that combination.