Anthropic has told investors it expects to post its first profitable quarter in June, after a sharp revenue acceleration outpaced its huge compute spending.
The AI research lab has projected revenue of $10.9 billion in the second quarter of 2026, doubling its first-quarter revenue. That is also more than the company’s annual revenue run rate of $9 billion, which it reached at the end of 2025. This has resulted in $559 million in operating profit for Anthropic, according to reporting by the Financial Times.
Anthropic is the first of three big AI companies looking to go public this year — the other two being OpenAI and SpaceX — to reach any form of profitability. While SpaceX was profitable pre-merger, it has been hampered by xAI’s huge compute spend. OpenAI has not signaled it is anywhere near profitability, with some questioning whether it will ever reach that point.
Anthropic leading in the enterprise market
Anthropic has been on a run of strong product announcements and has become the preeminent platform for AI-using programmers. At the same time, Mythos’s sophistication in identifying vulnerabilities and bugs may make it indispensable to governments, financial firms, and other large-scale enterprises, potentially opening the door to even larger contracts in the future.
Its tiff with the Pentagon boosted its support in the consumer market, but it still does not have anywhere near the scale of OpenAI or Google Gemini, both of which have 900 million active users.
However, it has become clear that the enterprise market is the more lucrative of the two at the moment. Claude has become a more popular option for businesses looking for AI tools that can support coding, research, customer operations, and other knowledge-work tasks without leaning as heavily on consumer adoption.
OpenAI has focused recent product announcements on the enterprise market, while Google spent a large portion of its I/O conference unveiling programs specifically for agentic coding and enterprise customers.
Alongside AI developers repositioning toward the enterprise market, investors are also seeing the value. Anthropic is reportedly close to closing a $30 billion funding round that would value it at $900 billion, slightly higher than OpenAI’s $852 billion valuation.
More must-read AI coverage
Huge compute spend expected over the next few years
Compared to the other two, Anthropic has been more careful with its compute spend.
OpenAI has said it will spend $600 billion on AI infrastructure by 2030 to build out its capacity, and has signed deals with SoftBank and Oracle to build 10 gigawatts of capacity through its Stargate project. xAI reportedly posted a net loss of $6.4 billion last year on $3.2 billion in revenue, although that may be much lower this year as the company sells excess capacity to rivals, including Anthropic, to shore up its finances.
Still, it would be surprising if Anthropic continued to post profitable quarters. It has had the worst outages of the three and has recently received $25 billion from Amazon and $40 billion from Google, with stipulations that some of it be spent on their cloud computing.
Even if Anthropic posts a profitable quarter, the harder test will be whether it can repeat the feat as model training, cloud commitments, and enterprise competition intensify. For now, the reported profit projection gives Anthropic a cleaner investor story than many of its AI rivals: growth that may finally be outrunning the compute bill.
For more on Anthropic’s enterprise AI momentum, read our coverage of Andrej Karpathy joining the company.
Read the full article here
