Anthropic is moving towards its first quarterly operating profit as corporate demand for Claude accelerates, marking a rare financial milestone for a frontier artificial intelligence lab still facing heavy spending on chips, data centres and model development. The San Francisco company has told investors that sales for the June quarter could reach at least $10.9 billion, more than double the $4.8 billion recorded in the March quarter, with operating profit expected at about $559 million. The figures point to a sharp shift in the economics of generative AI. For much of the past two years, the sector has been defined by vast losses, aggressive hiring and infrastructure commitments running into tens of billions of dollars. Anthropic’s expected profit suggests that enterprise adoption, particularly among developers, financial services firms, consultancies and regulated industries, is beginning to offset the cost of serving advanced models at scale.
Claude has gained traction with businesses looking for AI systems that can handle software engineering, document analysis, cybersecurity workflows and internal knowledge tasks. Anthropic said earlier this year that customers spending more than $100,000 annually on Claude had grown sevenfold over the past year, while more than 500 customers were spending above $1 million on an annualised basis. Eight of the Fortune 10 were Claude customers, underscoring the company’s growing presence among the world’s largest corporations.
Developer adoption has become one of the company’s strongest revenue engines. Claude Code, made publicly available in May 2025, has reached more than $2.5 billion in run-rate revenue, more than doubling since the beginning of 2026. Weekly active users have doubled since January 1, while business subscriptions have quadrupled. Enterprise use now accounts for more than half of Claude Code revenue, reflecting the growing willingness of companies to pay for AI tools that can write, debug and maintain software.
A major test for Anthropic will be whether that momentum can survive the next phase of infrastructure spending. SpaceX disclosed that Anthropic had agreed to pay $1.25 billion a month through May 2029 for access to AI compute capacity at the Colossus and Colossus II data centre clusters. Either side can terminate the agreements with 90 days’ notice, while fees are reduced during the initial capacity ramp-up period.
The arrangement highlights both the opportunity and the strain facing leading AI developers. Compute capacity has become a strategic bottleneck as companies race to train larger models and serve high-volume enterprise workloads. Anthropic’s ability to turn a quarterly operating profit while committing to major infrastructure spending gives it a stronger narrative with investors, but the company could return to losses if training costs rise, customer growth slows or pricing pressure intensifies.
Corporate partnerships are also widening Claude’s distribution. KPMG and Anthropic announced a global alliance on May 19 to embed Claude into KPMG’s Digital Gateway platform, starting with tax and legal services and expanding across client work. KPMG’s workforce of more than 276,000 people will gain access to Claude, and Anthropic has named the firm a preferred consultant for private equity work.
The KPMG agreement shows how frontier AI firms are trying to move beyond consumer subscriptions and developer tools into deeply embedded enterprise systems. Professional services groups are attractive partners because they handle high-value, document-heavy work across tax, legal, advisory and due diligence functions. At the same time, the shift raises questions about governance, accountability and the degree to which AI systems should influence client-facing professional judgement.
Anthropic’s financial position has been strengthened by a large capital raise. The company announced a $30 billion Series G funding round in February at a $380 billion post-money valuation. The fundraising reflected confidence in Claude’s commercial momentum, but it also underlined the scale of capital required to compete with OpenAI, Google, Meta and Elon Musk-linked AI ventures.
The company’s pitch rests partly on safety and reliability. Anthropic was founded by former OpenAI employees and has positioned Claude as a model family designed for steerability, enterprise controls and higher trust in sensitive workflows. That positioning has helped it win customers in sectors where accuracy, confidentiality and compliance matter, although the broader AI industry continues to face scrutiny over hallucinations, copyright disputes, labour disruption and energy use.
Competition remains intense. OpenAI continues to dominate consumer awareness through ChatGPT and has been expanding enterprise products. Google is pushing Gemini across Workspace, Cloud and Android. Meta has focused on open-weight models and developer distribution, while xAI is using Musk-linked infrastructure to compete aggressively in model training. Anthropic’s challenge is to preserve its enterprise momentum while avoiding dependence on a narrow set of large customers and costly compute providers.
Topics
Technology