Anthropic, the company behind the Claude AI model, expects to post its first profitable quarter.

The revenue ramp that made it possible

Anthropic’s annualized revenue sat at roughly $1B back in December 2024. By early 2026, that figure had climbed to approximately $30B in annualized run rate, with some estimates pegging the number as high as $43B by April 2026.

The growth has been driven largely by enterprise customers who have flocked to Claude for coding assistance, research, and complex reasoning tasks.

Spending big, but spending smarter

Anthropic’s training costs are projected to peak around $30B. Compare it to OpenAI’s projected $121B spend by 2028, and the picture shifts dramatically. Anthropic is building competitive models at roughly a quarter of the cost.

The company had previously set internal targets of reaching cash flow positive status by around 2027, with full profitability expected between 2028 and 2029. Hitting a profitable quarter now suggests those timelines were conservative.

A war chest that bought time, and leverage

Anthropic raised $13B in a Series F round that valued it at $183B, with ICONIQ leading the deal. It then followed up with a massive $30B Series G round at a $380B valuation, led by GIC and Coatue.

What this means for investors and the AI landscape

With training costs projected at roughly a quarter of OpenAI’s, Anthropic has demonstrated that research efficiency can substitute for brute-force spending.

The risk is that profitability in one quarter doesn’t guarantee sustained profitability. AI model training costs are lumpy. A quarter where Anthropic isn’t actively training a new frontier model will naturally look better on the expense side.