In mid-May, two AI giants revealed their cards—OpenAI secretly filed for an IPO, while Anthropic provided its first quarterly financial forecast.
The data shows that OpenAI had a revenue of $5.7 billion in the first quarter, but for every $1 earned, it lost $1.22. During the same period, Anthropic had a revenue of $4.8 billion, lagging behind by nearly $1 billion. However, the forecast for the second quarter shows a staggering growth rate, projected to reach $10.9 billion in revenue, with an operational profit of approximately $559 million.
This difference gives the impression to the public that one is a superstar valued at over a trillion dollars, still asking the market for patience. The other is a former follower who has quietly reached the profitability threshold.
Insiders revealed to The Information that OpenAI generated approximately $5.7 billion in revenue in the first quarter of this year, a number that is nearly $1 billion higher than the $4.8 billion revenue of its longtime rival Anthropic during the same period. Simply looking at these two numbers, OpenAI’s lead seems quite evident.
The aforementioned insiders revealed that three main factors drove OpenAI’s growth in the first quarter: the explosive popularity of the Codex programming AI, enterprise sales growth, and ChatGPT ad testing. The explosion of Codex indicates a strong demand from the developer community for tools that can directly assist in their work, which actually overlaps with Anthropic’s customer base. The foray into ad business exposed OpenAI’s anxiety about finding a monetization outlet within its vast pool of free users.
In the first quarter, OpenAI had an average of about 905 million weekly active users, peaking at 920 million in February. As the user base reaches an extremely high scale, growth begins to stagnate. Despite having 55 million paid consumer subscription users, an increase from the 47 million at the end of last year, the conversion rate remains low compared to its over 900 million weekly active users. Moreover, the corresponding inference cost is also a massive black hole for OpenAI.
On the other hand, Anthropic generated $4.8 billion in revenue in the first quarter, almost entirely from its core competency, selling AI models to enterprises and developers. It does not have a large free consumer base like ChatGPT that requires substantial subsidies. This difference may be a key element for it to surpass its old rival in the future.
According to financial data disclosed by Anthropic to investors as reported by The Wall Street Journal, the company expects second-quarter revenue to reach $10.9 billion, more than doubling from the first quarter. Furthermore, its revenue growth rate has already exceeded that of Google and Facebook pre-IPO.
The Information points out that by April 2026, Anthropic’s annualized revenue will exceed $300 billion, while OpenAI’s annualized revenue is around $250 billion. At a developer conference in May 2026, Anthropic CEO Dario Amodei joked that their recent revenue growth has been so fast that it is becoming “hard to handle.”
Anthropic is expected to achieve approximately $559 million in operating profit in the second quarter, which is also a milestone event. Last summer, the company shared a forecast with investors that they would not achieve full-year profitability until at least 2028. However, operating profit excludes equity incentive expenses, and considering the substantial ongoing computing expenses, Anthropic may not be able to maintain profitability throughout the entire fiscal year. But it has proven a point: AI model companies with enterprise customers at their core can establish a profitable model in a short period.
On the other hand, OpenAI, although its Q2 earnings are not yet known, data presented to investors shows that the company had an adjusted operating profit margin of -122% in Q1. In other words, for every $1 of revenue generated, it incurred a loss of $1.22. OpenAI expects to achieve positive cash flow in 2029 or 2030. Until then, it needs to continuously fill a significant funding gap.
HSBC analysts estimate that OpenAI faces a $207 billion funding gap relative to its growth plans. OpenAI CEO Sam Altman hinted at a company all-hands meeting that even after submitting IPO paperwork, the actual listing might be delayed because filing for an IPO and being “ready to go public are two different things.” The financial pressure behind this is self-evident.
Why would two companies see such a drastic difference in their financial situations in the same AI wave? The answer lies in their different customer structures. According to Forbes’ analysis, Anthropic receives about 85% of its revenue from enterprise and developer customers. Over 500 companies already spend over $1 million annually on the Claude platform, with 8 of the Fortune 10 companies being its customers.
Enterprise customers have a clear willingness to pay, a more predictable query pattern, lower service costs, and more sticky contracts. This is a healthy, sustainable business model. For Anthropic, in Q1, each $1 earned in revenue required $0.71 to be spent on computing power. By Q2, this number is expected to drop to $0.56, showing immediate efficiency gains.
Conversely, OpenAI relies on ChatGPT consumer subscriptions for about 85% of its revenue. While it has 55 million paying subscribers, behind them are over 900 million weekly active users with no corresponding revenue to cover, leading to a structural deficit. OpenAI is not unaware of this. Under the leadership of CEO of Applied AI, Fidji Simo, and other executives, the company has begun to cut money-burning projects like the video generation app Sora, attempting to shift focus towards revenue-generating businesses and commercial customers. However, changing a business model centered around free consumers is not an overnight task.
Of course, directly comparing the revenue figures of two companies requires taking into account a key accounting treatment difference. The Information explained this in detail: Anthropic recognizes all revenue from technology sales through cloud partners such as Amazon and Google as full revenue. On the other hand, due to its long-term special partnership with Microsoft where Microsoft has proprietary rights to use its intellectual property, OpenAI only recognizes 20% of the revenue from model sales through Microsoft Azure as its own revenue.
However, it is important to note that there are slight accounting differences between the two companies, and both have some degree of “revenue inflation”: Anthropic records the full revenue from reselling its models through cloud providers like Amazon and Google without deducting the revenue share; OpenAI, on the other hand, needs to share 20% of its revenue with Microsoft before 2030 (which could reach $6 billion this year), and does not account for sales generated through cloud partners in its disclosures. Even if OpenAI were to adopt Anthropic’s accounting method to increase its annualized revenue by several billion dollars, it would still not be able to bridge the already tens of billions of dollars gap between the two.
On the IPO path, all financial secrets will be laid bare. OpenAI, Anthropic, and Elon Musk’s SpaceX are all racing to go public, with valuations of all three companies potentially exceeding $1 trillion.
Currently, OpenAI has raised $122 billion in funding from suppliers such as Amazon and NVIDIA and is seeking to go public as early as September 2026. Meanwhile, Anthropic is in the midst of a funding round that could value it above OpenAI and is considering an IPO as early as October. Ultraman privately expressed his desire to go public first. Anthropic currently holds a quarter with proven profitability.
Even if future losses occur due to astronomical-scale investments in computing infrastructure, such as paying $1.25 billion monthly to SpaceX for leasing data center capacity, and new mega-deals with Broadcom and Google for computing power, it has once again shown the market that its business model is viable. Its story is one of an enterprise software company that can be compared to Salesforce or ServiceNow.
OpenAI, on the other hand, presents a story that requires stronger conviction from public market investors. It needs to convince the market to believe that AI agents, image generation, and its future vast advertising business will eventually turn its massive consumer traffic into profit. In Ultraman’s roadmap, by 2030, ChatGPT’s advertising business could bring in about $1020 billion in revenue. But that will take time, and time happens to be the scarcest thing for OpenAI as it trades losses for growth.
OpenAI has just brought online over 2 gigawatts of compute, exceeding the entire Colossus cluster of SpaceX, all of which costs money. So for investors, when the S-1 document is made public, should we believe in a company that has already found a profitable model, or in a behemoth that asks the market to give it several more years and hundreds of billions of dollars to explore profit possibilities? The answer will determine the fate of both companies.
[BlockBeats]
AI Giants’ Diverging Paths: Profitability vs. Growth and Implications for Crypto
The recent financial disclosures from OpenAI and Anthropic reveal a fascinating divergence in the AI landscape that carries significant implications for the crypto market. While OpenAI continues to burn cash with a staggering -122% operating profit margin, Anthropic appears to have quietly achieved quarterly profitability, highlighting crucial business model differences that crypto investors should heed.
Financial Disparity: Two AI Titans, Different Trajectories
The financial contrast between these AI behemoths is striking. OpenAI reported $5.7 billion in Q1 revenue but incurred $1.22 in losses for every dollar earned. In contrast, Anthropic, with slightly lower Q1 revenue of $4.8 billion, projects Q2 revenue of $10.9 billion with an operating profit of approximately $559 million. More remarkably, Anthropic anticipates reaching $300 billion in annualized revenue by April 2026, surpassing OpenAI’s projected $250 billion.
This divergence stems fundamentally from their customer structures. Anthropic derives approximately 85% of its revenue from enterprise and developer customers, with over 500 companies spending at least $1 million annually on its Claude platform, including 8 of the Fortune 10 companies. OpenAI, conversely, relies on ChatGPT consumer subscriptions for about 85% of its revenue, despite having over 900 million weekly active users with a low conversion rate to paid subscriptions.
Enterprise Model Validation: A Crypto Precedent
Anthropic’s success validates what many Web3 projects have long believed: enterprise-focused business models often prove more sustainable than consumer-centric approaches. Enterprise customers offer clearer willingness to pay, more predictable query patterns, lower service costs, and stickier contracts—creating a healthier, more sustainable revenue stream.
This dynamic should inform crypto investment strategies. Projects targeting enterprise clients with clear ROI propositions may demonstrate stronger fundamentals than those pursuing mass consumer adoption. The efficiency gains Anthropic has achieved—reducing computing costs from $0.71 to $0.56 per dollar of revenue—highlight the importance of optimizing unit economics, a critical consideration for any blockchain-based business.
Token Implications: Valuing Profitability Over Growth
The contrasting approaches of OpenAI and Anthropic reflect a broader shift in market sentiment toward valuing profitability over unrestrained growth. For crypto investors, this suggests that projects with demonstrated revenue models and clear paths to profitability may be better positioned for long-term success than those prioritizing user growth at all costs.
AI-focused crypto tokens should take note. Projects like SingularityNET (AGI), Fetch.ai (FET), and Ocean Protocol (OCEAN), which have pursued enterprise partnerships, may find their strategies validated by Anthropic’s success. Conversely, tokens reliant on speculative consumer growth without clear monetization paths may face increasing skepticism from investors.
Infrastructure Battles: Compute as the New Digital Real Estate
Both AI giants are engaging in massive infrastructure spending, with OpenAI recently activating over 2 gigawatts of compute capacity—exceeding SpaceX’s entire Colossus cluster. Anthropic is paying $1.25 billion monthly to SpaceX for data center capacity while securing new mega-deals with Broadcom and Google.
This creates significant opportunities for decentralized infrastructure solutions in the crypto space. Projects offering alternative compute resources—like Render Network (RNDR), Akash Network, or Bittensor—may benefit as centralized AI providers face rising costs and potential bottlenecks. The tokenization of compute resources could emerge as a compelling value proposition, offering more efficient allocation mechanisms than traditional cloud infrastructure.
Accounting Nuances and Market Transparency
The article highlights important accounting differences between the two companies. Anthropic recognizes full revenue from cloud partners, while OpenAI only recognizes 20% of Microsoft Azure revenue due to their partnership structure. These nuances underscore the importance of transparency in financial reporting—a principle crypto projects would do well to embrace as they mature.
For crypto investors, this reinforces the need for rigorous due diligence, particularly when comparing projects with different tokenomic structures or revenue recognition models. The coming public listings of both AI companies will provide additional transparency that could reshape market expectations.
The Crypto-AI Convergence: Strategic Opportunities
The competitive dynamics between OpenAI and Anthropic reveal several strategic opportunities for the crypto sector:
-
Decentralized AI Networks: The challenges faced by centralized AI providers create opportunities for decentralized alternatives that offer greater transparency, efficiency, and user control. Projects that effectively leverage blockchain to address AI’s current limitations could capture significant value.
-
Data DAOs: The demonstrated value of high-quality training data for AI models suggests that decentralized data markets could become increasingly important. Projects enabling community-owned data ecosystems may benefit from the growing AI demand.
-
Tokenized AI Services: As AI becomes more valuable, new models for accessing these services through tokens could emerge. This could create entirely new asset classes within the crypto ecosystem.
-
Cross-Chain AI Solutions: The resource-intensive nature of AI development may drive demand for solutions that can leverage multiple blockchain networks for optimal efficiency and cost-effectiveness.
Investment Considerations for a Maturing Market
As the AI market matures and these companies prepare for IPOs, crypto investors should consider several factors:
-
Risk Profile: OpenAI’s need for $207 billion in additional funding highlights the risks of growth-at-all-costs strategies. Crypto projects with more balanced approaches to growth and profitability may be more resilient in market downturns.
-
Market Timing: The race to between OpenAI and Anthropic to go public suggests the AI market is approaching a critical inflection point. Crypto projects that can demonstrate clear value propositions before this wave crestes may benefit from increased institutional interest.
-
Token Utility: The success of AI services with clear commercial applications suggests that tokens with tangible utility—rather than purely speculative value—may perform better in the long term.
-
Regulatory Considerations: As AI companies face increased scrutiny, crypto projects operating in this space should proactively address regulatory concerns to avoid potential headwinds.
Conclusion: The Path to Sustainable Value Creation
The diverging trajectories of OpenAI and Anthropic offer valuable lessons for the crypto market. Anthropic’s success demonstrates that enterprise-focused models with clear monetization paths can achieve profitability more quickly than consumer-centric approaches that rely on massive user bases without corresponding revenue.
For crypto investors, this suggests a shift toward valuing projects with:
– Enterprise partnerships with clear ROI
– Demonstrable unit economic efficiency
– Transparent tokenomics aligned with real economic activity
– Sustainable infrastructure solutions
As the AI and crypto convergence accelerates, the most successful projects will likely be those that can effectively leverage blockchain to address the limitations of centralized AI providers while creating new forms of value that benefit all participants. The coming years will likely see the emergence of a more sophisticated crypto market that prioritizes sustainable value creation over speculative growth.