New analysis reveals that Anthropic is redefining AI commercialisation with profitability and growth rates that significantly outpace its rivals. Leveraging a high-margin, API-centric business model, the company has established itself as a leader in the B2B AI market.

According to a detailed report, Anthropic is projected to achieve a GAAP EBIT of $1 billion in the third quarter of 2026, representing a margin of approximately 6%. Concurrently, its Annual Recurring Revenue has surged from $9 billion at the end of 2025 to over $60 billion presently. The analysis forecasts that if Anthropic maintains its current pace of roughly $15 billion in monthly net new ARR, its ARR could reach $300 billion by the end of 2027, implying an enterprise value of $6 trillion and positioning it as the world's most valuable company.

Anthropic confidentially filed for an initial public offering on June 1st. The analysis suggests the timing is strategically urgent, as capital markets are tightening with competitors like Alphabet completing substantial equity financing rounds and Meta reportedly planning a multi-billion dollar raise. The report contends that Anthropic's superior financial metrics and business model mean it should aim to go public ahead of OpenAI to seize the initiative in the capital competition.

Claude Code Ignites B2B Market, Driving Over 200% Quarterly ARR Growth

Anthropic's performance inflection is attributed to the explosive adoption of Claude Code. Data indicates Claude Code now accounts for over 7% of all code commits on GitHub, directly propelling the company's monthly ARR additions from $3 billion in January to $11 billion in March within the first quarter.

The revenue structures of Anthropic and OpenAI show a stark divergence. Approximately 75% to 85% of Anthropic's ARR comes from usage-based API billing, with consumer subscriptions constituting only about 5% of total ARR. In contrast, over 65% of OpenAI's revenue in Q1 2026 still stemmed from subscription models, with consumer ARR making up around 40%.

The analysis highlights the core advantage of the API model: the absence of a per-user revenue ceiling. As existing customers adopt more agentic workflows, their token consumption and corresponding revenue grow continuously, enabling expansion without acquiring new clients. Anthropic's Chief Financial Officer disclosed in a May podcast that the company's net revenue retention rate stands at an extraordinary 500%. This means customers who contributed $30 billion in ARR in the first quarter were only contributing $2 billion a year prior.

Gross Margin Advantage Fuels Compounding Flywheel, OpenAI Lags

The difference in business models is directly reflected in gross margins. Estimates suggest Anthropic's current consolidated gross margin has risen to a range in the mid-60s percent, a dramatic improvement from negative 94% in 2024. Its API business gross margin exceeds 80%.

The primary driver for this margin expansion is improved inference efficiency. Measured by ARR per megawatt of compute power, Anthropic's metric is projected to reach $60 million later this year, up from just $16 million nine months ago. Since inference compute costs are largely fixed, marginal profit approaches 100% when more tokens are processed per unit of compute or when token pricing increases.

The report calculates that if both Anthropic and OpenAI were to reach $100 billion in ARR, OpenAI's gross profit would be approximately $25 billion lower than Anthropic's due to the cost of supporting over 900 million free users. This gap would directly impact each company's capacity to reinvest in next-generation model training.

The analysis introduces "Earnings Before Training and Interest Taxes" as a key metric for measuring a lab's reinvestment capacity, with Anthropic achieving a 36% EBTIT margin in Q2 2026. It predicts that by 2028, Anthropic's cumulative EBTIT will exceed OpenAI's by $250 billion.

Beyond Programming, Cybersecurity Emerges as Next Growth Engine

Current estimates indicate over 65% of the lab's ARR originates from programming-related use cases, with coding tool startups collectively contributing about $6 billion in ARR. Meta is Anthropic's largest single customer, but its contribution remains between 3% and 5%.

The report identifies cybersecurity as the next explosive vertical following programming. It anticipates that the release of the new Fable model will further increase token pricing, expand use cases, and push monthly net new ARR in the second half of 2026 beyond the current $10 billion per month level. Healthcare, finance, and biotech are also listed as verticals with significant potential for total addressable market expansion.

Regarding distribution channels, the indirect "Token-as-a-Service" model sold through hyperscale cloud platforms like AWS Bedrock and Azure Foundry is growing rapidly. It now accounts for 15% to 20% of Anthropic's ARR, up from just 5% to 10% a quarter ago. The analysis argues that paying a 20% to 30% revenue share to these platforms remains economically justifiable given the efficiency in reaching enterprise clients and compliance benefits.

Compute Bottleneck is Key Variable, IPO Provides Funding Pathway

The core constraint on Anthropic's growth prospects is compute supply.

Forecasts suggest that by 2030, the combined unconstrained compute demand from Anthropic and OpenAI will exceed 100 gigawatts. However, net new compute capacity for 2025 and 2026 is only 2.5 GW and 5 GW respectively, with the two companies' current combined available compute standing at just over 6 GW.

This supply-demand gap underscores the strategic importance of the IPO. The report states that funds raised from the public offering will be primarily used to bridge the widening gap between compute needed for inference operations and new model training, allowing the company to lock in compute resources at more favourable financing costs. It also mentions market speculation that Meta is considering renting out compute to external parties, with an expectation that Anthropic would procure incremental compute from such trusted suppliers.

The analysis also lists key risk factors, including rumoured price cuts from OpenAI, competitive pressure from Google DeepMind and Meta in programming models, potential government regulatory restrictions on frontier model releases, and the dilutive effect on consolidated gross margins from a rising share of TaaS revenue. The report explicitly states that if regulatory hurdles impede model releases and narrow the capability gap between open-source and frontier proprietary models, it would fundamentally erode Anthropic's commercial moat.