Anthropic is on track to reach $1 billion in GAAP operating profit by the third quarter of 2026, according to new figures from SemiAnalysis, marking a sharp financial turn for one of the most closely watched artificial intelligence companies as it prepares for a public listing.
The research firm said Anthropic’s annual recurring revenue, or ARR, has climbed from $9 billion at the end of 2025 to more than $60 billion. If the company keeps expanding at its current pace, SemiAnalysis estimates ARR could reach $300 billion by the end of 2027, a level that could imply an enterprise value close to $6 trillion.
The projections place Anthropic at the center of a fast-changing AI market where revenue growth, computing capacity and access to capital are becoming increasingly connected. The company filed for an initial public offering on June 1, a move described in the report as strategically urgent as funding conditions tighten and rivals raise enormous sums to secure chips, data centers and power.
The IPO filing comes as large technology companies intensify their race to fund AI infrastructure. Alphabet has raised $84.75 billion in new equity, while Meta is reportedly preparing multi-billion-dollar funding rounds. For Anthropic, the central issue is no longer only demand for its models, but whether it can obtain enough computing resources to serve that demand at acceptable margins.
The company’s rapid growth has been driven largely by Claude Code, its AI coding tool, which SemiAnalysis said now accounts for more than 7% of all GitHub submissions. That adoption helped quarterly monthly ARR additions rise from $3 billion in January to $11 billion in March, showing how quickly developer usage has become a core engine of the company’s revenue base.
Revenue growth is being driven by usage
Anthropic’s business model differs sharply from subscription-heavy AI platforms. SemiAnalysis estimated that 75% to 85% of Anthropic’s revenue comes from usage-based API services, while consumer subscriptions contribute about 5%. OpenAI, by comparison, gets more than 65% of its income from subscriptions, according to the report.
That difference matters because Anthropic’s revenue growth is tied more directly to the amount of computing its customers consume. Instead of relying mainly on fixed monthly payments, the company benefits when developers, enterprises and software platforms increase token usage across its systems.
Chief financial officer Rao reported a net revenue retention rate of 500%, according to SemiAnalysis. In practical terms, the report said customers that generated $2 billion in ARR a year ago contributed $30 billion this quarter. Such a figure suggests Anthropic can grow rapidly even without a matching increase in its customer count, as existing users expand activity across its models.
This structure gives traders and analysts a clearer way to measure demand for AI deployment. Rising revenue is not only a sign of customer acquisition; it reflects deeper integration of Anthropic’s systems into software development, enterprise workflows and automated services.
Programming is currently the company’s strongest market. SemiAnalysis estimated that about 65% of Anthropic’s current ARR comes from programming applications. Developer tool startups together contribute about $6 billion, while Meta remains Anthropic’s largest customer, accounting for roughly 3% to 5% of revenue.
The scale of coding-related demand also highlights why Claude Code has become so important to Anthropic’s financial outlook. As AI-assisted programming becomes more common inside businesses and developer platforms, usage can rise quickly without requiring the slower sales cycles often associated with traditional enterprise software.
Margins have improved sharply
Anthropic’s financial profile has changed dramatically over the past year. SemiAnalysis said the company’s overall gross margin has climbed into the mid-60% range, compared with negative 94% in 2024. API margins are now above 80%, reflecting better efficiency, higher utilization and stronger pricing power in usage-based services.
Efficiency gains are also visible in the company’s revenue per unit of computing power. ARR per megawatt of computing capacity has increased from $16 million nine months ago to an expected $60 million later this year, according to the report. That improvement is critical because power and hardware availability are now among the biggest constraints in the AI sector.
SemiAnalysis introduced a metric called “earnings before training and interest and taxes,” meant to separate the economics of serving customers from the heavy cost of training new models. On that basis, the firm placed Anthropic’s second-quarter 2026 margin at 36%.
The report also estimated that Anthropic could build a cumulative advantage of $250 billion by 2028 under this framework. The advantage is tied partly to its smaller consumer-facing burden compared with OpenAI. At an equivalent $100 billion ARR level, SemiAnalysis estimated that OpenAI’s need to serve about 900 million free users would reduce gross profit by $25 billion compared with Anthropic.
That comparison underscores a central debate in the AI industry: whether the strongest financial model is built around broad consumer reach, enterprise API usage, developer workflows or some mix of all three. Anthropic’s current numbers suggest that heavy usage from businesses and software builders can produce high-margin growth if computing costs are kept under control.
The IPO is tied to the computing race
Anthropic’s planned public offering is more than a conventional liquidity event. It is part of a broader race to secure the physical infrastructure needed to run advanced AI models.
The company’s funding needs are rising because demand for compute is growing faster than available supply. SemiAnalysis forecast that the combined unconstrained power needs of Anthropic and OpenAI could exceed 100 gigawatts by 2030. Current available capacity is just over 6 gigawatts, with planned additions of 2.5 gigawatts in 2025 and 5 gigawatts in 2026.
That gap shows why AI companies are competing not only for customers but also for electricity, chips, data-center space and long-term supply contracts. Even companies with strong revenue growth may be limited by how much computing capacity they can access.
Funds raised through Anthropic’s impending IPO are expected to help cover rising infrastructure costs and secure computing supply at lower financing rates. A public listing could also give the company more flexibility as it negotiates long-term agreements with cloud providers, chip suppliers and infrastructure partners.
Market reports from early July suggested Meta may lease unused capacity, with Anthropic likely to seek additional resources from such partners. Such arrangements could help bridge near-term supply gaps, though they may also create dependencies on large technology companies that are both partners and competitors in the AI race.
Cloud distribution is gaining importance
Sales through large cloud providers are becoming a larger part of Anthropic’s business. SemiAnalysis said cloud-based “token-as-a-service” distribution now accounts for 15% to 20% of total ARR, up from 5% to 10% in the previous quarter.
These arrangements allow customers to access Anthropic’s models through familiar cloud platforms. Revenue-sharing rates of 20% to 30% are viewed as workable because cloud partners can provide customer relationships, compliance support and distribution scale.
Cloud distribution can accelerate growth, especially among large enterprises that prefer to buy AI services through existing procurement channels. It can also help Anthropic reach companies that require strict compliance, data governance and security controls.
However, the model may pressure margins over time. Sharing revenue with cloud providers reduces the amount Anthropic keeps from each transaction, even if the channel expands total usage. The company’s challenge will be to use cloud platforms for scale without giving up too much pricing power or customer control.
Cybersecurity could become the next major market
Programming remains Anthropic’s dominant use case, but SemiAnalysis identified cybersecurity as the next major growth area. The launch of the Fable model could support higher token pricing and new applications in security operations, threat detection and automated response.
AI tools are increasingly being tested for tasks such as code review, vulnerability detection, incident analysis and system monitoring. If those use cases mature, cybersecurity could become a natural extension of Anthropic’s developer-focused business.
Healthcare, finance and biotechnology were also cited as potential demand drivers. These industries often handle complex data, intensive documentation and high-value workflows, making them attractive markets for advanced AI systems. They also face stricter regulatory and compliance standards, which may favor providers that can demonstrate reliability and control.
For Anthropic, expansion into these sectors could diversify revenue beyond coding tools. Still, success in regulated industries usually requires longer sales cycles, stronger governance and careful management of model risk.
Risks remain despite rapid expansion
SemiAnalysis identified several risks that could affect Anthropic’s outlook. OpenAI could reduce prices, putting pressure on usage-based revenue across the market. DeepMind and Meta could become stronger competitors in coding models, especially if they improve performance or use open-source strategies to gain adoption.
Regulation is another major uncertainty. Tighter oversight of advanced AI systems could slow model releases, raise compliance costs or limit the deployment of new capabilities. If rules narrow the performance gap between proprietary systems and open-source alternatives, Anthropic’s competitive position could weaken.
Margin pressure from cloud-based distribution is also a concern. As more revenue flows through large cloud platforms, Anthropic may need to accept lower direct revenue per token in exchange for broader reach. That trade-off can support growth but may reduce profitability if infrastructure costs rise at the same time.
The report concluded that barriers to entry could narrow if regulation restricts new model releases or if open-source systems close the performance gap with leading proprietary models. In that scenario, pricing power could decline and customers might gain more leverage.
A test for capital markets
Anthropic’s expected public offering comes at a moment when the AI sector’s financial ambitions are colliding with physical constraints. The company’s revenue growth has been extraordinary, but the next stage depends on power, chips, data centers and financing.
The broader market impact could be significant. A large AI IPO may absorb substantial liquidity, especially if other major AI-related companies pursue funding at the same time. Traders are likely to watch whether demand for Anthropic shares pulls capital away from other high-growth technology sectors.
The offering may also influence how public markets value AI companies. Traditional software metrics such as subscription growth and customer count may be less useful for companies whose revenue is driven by token consumption and computing intensity. Anthropic’s reported 500% net revenue retention rate points to a model where existing customers can dramatically increase spending as they automate more work.
At the same time, the costs are unusually physical for a software company. Data centers require land, electricity, cooling systems, chips and long-term power arrangements. That makes Anthropic’s valuation partly a bet on its ability to secure infrastructure ahead of demand.
Infrastructure may decide the next phase
The most durable theme in Anthropic’s story is the infrastructure bottleneck. AI demand is growing faster than the power systems and chip supply chains needed to support it. If SemiAnalysis’ forecast of more than 100 gigawatts of combined unconstrained power demand for Anthropic and OpenAI by 2030 proves close to accurate, the industry will need a massive build-out of energy and computing capacity.
That reality puts semiconductor makers, cloud providers and power suppliers at the center of the AI economy. The companies that can provide reliable access to advanced chips and electricity may become as important as the model developers themselves.
For Anthropic, the IPO is intended to strengthen its position in that race. The company has moved from steep negative margins in 2024 to projected GAAP operating profit in 2026, while ARR has expanded at a pace rarely seen in enterprise technology. But continued growth will depend on whether it can turn demand into delivered service without running into capacity shortages or cost inflation.
The coming public offering will therefore test more than appetite for a fast-growing AI company. It will test whether public markets are ready to fund the next phase of AI infrastructure at the scale required. Anthropic’s numbers suggest enormous commercial momentum, but the company’s future may be shaped as much by megawatts and financing terms as by model performance.
For deeper insight into AI’s impact on finance and markets, explore our guide on AI in banking today.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
