Anthropic's internal Claude usage value exceeds monthly employee costs

Anthropic's SEC filing shows the list-price value of its internal Claude use first exceeded monthly personnel expenses in April 2026, while usage value rose more than 250-fold by September

Anthropic's internal use of its artificial intelligence (AI) assistant Claude has grown to a point where its estimated monthly value exceeds the company's expenses on workers, showing the intensity with which the technology is being deployed within the AI firm.

 

The value of Anthropic's internal Claude usage, calculated at the prices it charges customers, first exceeded its monthly personnel expenses, excluding stock-based compensation, in April 2026, the company disclosed in an amended initial public offering (IPO) filing with the US Securities and Exchange Commission (SEC).

 

The company also said the value of its internal Claude usage increased more than 250 times between January and September 2026.

Claude writes more than 90% of code contributions

Anthropic said more than 90 per cent of contributions to its codebase in September 2026 were written by Claude.

 

The average daily volume of code contributed to the company's codebase per engineer in the three months ended September 30, 2026, was more than 45 times the average daily volume recorded in 2024, according to the filing.

 

The company said Claude is increasingly being used to run experiments, generate and process training data, automate evaluations and safety audits, and write and review code. Its engineers are consequently spending more time directing and reviewing AI-generated work rather than writing code line by line.

 

Anthropic described this process as a feedback loop in which successive versions of Claude help accelerate the development of subsequent models.

 

The company also drew a comparison between its workforce and revenue growth. Its headcount increased more than fivefold between December 31, 2024, and September 30, 2026. However, revenue in the July-September 2026 quarter was 86 times the revenue recorded in the October-December 2024 quarter.

 

Anthropic attributed this operating leverage partly to the use of Claude across its research, engineering, sales and operations teams, saying employees could use multiple AI agents to undertake work that would traditionally require larger teams. 

 

“Because each version of Claude is used to help build the next and because Claude makes our teams more productive across the company, we believe the compute we devote to internal usage compounds these gains, continuously accelerating our pace of development and further strengthening our leadership position,” the company said in the filing.

Anthropic flags risks of AI developing AI

The company also acknowledged potential safety concerns arising from the growing role of AI in developing more capable AI systems.

 

Using AI to help build subsequent generations of models could pose a considerable safety risk, particularly if it accelerates the pace of technological progress, Anthropic said in the filing.

 

It said its Responsible Scaling Policy was designed to monitor and mitigate risks associated with what is known as recursive self-improvement, where AI systems contribute to the development of more advanced versions of themselves. 

 

“We believe our investments in safety techniques will enable us to manage these risks and responsibly realise substantially more leverage from Claude than we do today in developing AI models,” it added.