SAN FRANCISCO, United States — October 31, 2026
Anthropic, the artificial intelligence company behind Claude, has revealed extraordinary revenue growth ahead of a potentially record-breaking initial public offering. But its newly disclosed financial statements also expose the enormous cost of competing at the frontier of artificial intelligence.
The company generated $16.29 billion in revenue during the first six months of 2026, compared with just $1.16 billion during the same period a year earlier.
Yet its filing with the US Securities and Exchange Commission also reported an eye-catching $80.75-billion net loss.
The contrast is striking: an AI company growing at unprecedented speed while disclosing losses and financial obligations on a scale rarely seen in the technology industry.
However, the numbers tell a more complicated story than the headline loss suggests.
A substantial portion of Anthropic’s reported loss came from noncash accounting adjustments associated with financing instruments whose values increased as the company’s valuation rose.
Its operating loss, which more closely reflects the performance of its ongoing business, narrowed to approximately $693 million from $2.14 billion a year earlier.
At the same time, Anthropic spent almost $9 billion on computing and infrastructure during the first half of 2026.
That is the central dilemma investors will face when the company brings its shares to market.
Can Claude generate enough lasting revenue to pay for the extraordinary computing power needed to train and operate advanced AI—or will the industry’s relentless infrastructure race eventually overwhelm even its fastest-growing companies?
Anthropic Officially Files for a Potential Nasdaq IPO
Anthropic publicly filed its Form S-1 registration statement with the US Securities and Exchange Commission on October 26, 2026.
The preliminary prospectus provides one of the most detailed financial examinations yet of a leading independent AI model developer.
The company plans to list its Class A common shares on Nasdaq under the proposed symbol ANT.
But the filing is not the completion of an IPO.
Anthropic has not yet established its final offering price, number of shares to be sold or definitive market valuation.
Reports have discussed potential valuations ranging from approximately $1.5 trillion to $2 trillion, and estimates of how much the offering might raise have been extraordinarily large.
Those estimates remain market expectations, not finalized offering terms.
The proposed listing could become a critical test of investor demand for companies whose growth depends on huge investments in chips, electricity, data centers and cutting-edge AI research.
Revenue Jumps From $386 Million in 2024 to $16.29 Billion in Six Months
Anthropic’s revenue trajectory is among the most remarkable features of its filing.
| Reporting period | Revenue |
|---|---|
| Full year 2023 | $41 million |
| Full year 2024 | $386 million |
| Full year 2025 | $4.59 billion |
| First half of 2025 | $1.16 billion |
| First half of 2026 | $16.29 billion |
The latest six-month result represents approximately 14 times the revenue recorded during the comparable 2025 period.
The growth has been driven by increased business adoption of Claude models, AI coding tools and enterprise applications.
A large share of the company’s income comes from consumption-based products, where customers pay according to their use of AI models.
Anthropic also earns subscription revenue from paid products.
During the first six months of 2026, consumption-based revenue reached approximately $13.06 billion, while subscriptions contributed about $3.22 billion.
The figures demonstrate that Anthropic has developed a substantial commercial business rather than relying exclusively on investor funding.
But they also raise a difficult question: How much longer can such rapid expansion continue?
A company’s growth rate generally becomes harder to maintain as its revenue base becomes much larger.
Anthropic itself cautions that its previous extraordinary growth rates should not be assumed to continue indefinitely.
Why Anthropic’s $80.7-Billion Net Loss Is Misleading Without Context
The most sensational number in the IPO filing is the $80.75-billion first-half net loss.
Taken in isolation, that figure could suggest an operational business losing tens of billions in cash.
That interpretation would be inaccurate.
The company’s statements include approximately $72.4 billion in noncash charges associated with changes in the value of convertible financing instruments and related liabilities.
As Anthropic’s estimated enterprise valuation increased, the accounting value of certain obligations also increased.
Those changes produced large charges in the financial statements.
But these are not equivalent to Anthropic paying $72.4 billion in cash for cloud computing or employee salaries during the period.
The operating loss presents a very different picture.
Anthropic’s operating loss narrowed to $693 million in the first half of 2026, compared with $2.14 billion a year earlier.
On an adjusted, non-GAAP basis, the company reported approximately $1.62 billion in operating income.
This means Anthropic has made substantial progress toward operating profitability.
However, an adjusted operating profit is not the same as a GAAP net profit, and it does not eliminate the company’s substantial future financial commitments.
The 85% Gross Margin Looks Impressive — But It Doesn’t Tell the Whole Story
One of CNBC’s central questions concerns the surprisingly high gross margin suggested by the filing.
Anthropic reported $16.29 billion in first-half revenue and $2.38 billion in cost of revenue.
That produces an implied gross margin of approximately 85.4%.
On the surface, this resembles the economics of highly profitable software businesses.
But AI development introduces an important distinction.
The cost of serving customer requests is only one part of the expense of running a frontier AI company.
Anthropic also spends enormous amounts developing and training new models.
Those expenses are largely reflected elsewhere in the financial statements, particularly research and development.
For the first six months of 2026, Anthropic reported approximately $9.74 billion in research and development expenses.
Its total computing and infrastructure expenses reached about $8.95 billion.
Therefore, the company cannot be described as having an 85% profit margin.
The gross-margin calculation primarily measures the relationship between recognized revenue and costs allocated to delivering that revenue.
It does not capture every dollar spent on developing future versions of Claude.
Computing Power Is Anthropic’s Biggest Financial Challenge
Operating advanced AI models requires enormous computing resources.
These systems depend on specialized processors, high-speed networking, data-center facilities, electricity and sophisticated software infrastructure.
Anthropic uses a combination of computing technologies, including AWS Trainium chips, Google’s tensor processing units, Nvidia GPUs and AMD GPUs.
The company says its ability to move workloads across different computing systems helps optimize performance and cost.
But the level of investment required remains enormous.
In the first half of 2026 alone, Anthropic’s compute and infrastructure expenses reached approximately $8.95 billion.
The company has also entered into extensive contractual commitments with infrastructure and cloud partners.
Reuters reported that its commitments exceed $600 billion over the coming decade.
Other reporting examining the prospectus and associated agreements has described even larger potential obligations when additional arrangements are considered.
These figures should not automatically be interpreted as money already spent.
Contractual commitments, lease arrangements, potential expenditures and completed capital spending are different financial concepts.
Some agreements may involve substantial future payment obligations even if customer demand changes.
That makes the timing and profitability of future revenue particularly important.
The Hidden Question: Training AI Versus Serving Customers
One of the biggest uncertainties raised by CNBC is how much Anthropic spends on training future models compared with serving paying customers.
The distinction is essential to understanding the economics of artificial intelligence.
Serving customers involves using existing models to process requests.
Training involves the expensive process of developing new models or substantially improving their capabilities.
Anthropic’s filing provides some information about how expenses are allocated across different accounting categories.
But investors still lack a complete operational breakdown that would allow them to predict precisely how training costs will evolve.
Nick Patience, AI lead at the Futurum Group, told CNBC that understanding this allocation is essential for evaluating future margins.
If the cost of serving existing customers continues falling, Anthropic could generate greater gross profits as demand expands.
But if increasingly powerful models require vastly more spending on training, research and infrastructure, those gains could be offset.
That is why the company’s impressive gross margin does not settle the debate about its long-term profitability.
Amazon, Google and Microsoft Are Both Partners and Risks
Anthropic depends heavily on major cloud computing companies.
Amazon and Google are important providers of computing infrastructure and commercial distribution.
Both are also significant investors.
Microsoft has entered into major cloud arrangements with Anthropic.
This creates a complex business structure.
The same companies may supply computing power, distribute Claude products, invest in Anthropic and develop competing AI technology.
According to Reuters, approximately 47% of Anthropic’s 2025 revenue was routed through Amazon and Google’s cloud marketplaces.
That arrangement gives Anthropic access to enterprise customers already using major cloud platforms.
But it also creates concentration risk.
A change in pricing, infrastructure availability, commercial terms or strategic priorities at a major partner could affect Anthropic’s business.
And because several partners are also competitors, commercial relationships may become increasingly complicated.
The company identifies these dependencies as important risks in its IPO filing.
A $60-Billion AI Chip Financing Deal Shows How Expensive the Race Has Become
Anthropic’s financing strategy extends beyond conventional software investment.
WWC ONE MEDIA previously examined the issue in Anthropic Offers Blackstone More Than 10% Returns in $60 Billion AI Chip Deal — But the Rising Cost of AI Financing Signals a Bigger Problem.
That report discussed a proposed financing structure involving AI chips and substantial infrastructure spending.
The arrangement illustrates how institutional capital is being drawn into the AI computing market.
But expensive financing creates additional demands on future cash flows.
Lenders and infrastructure investors expect payment even if demand develops more slowly than anticipated.
For Anthropic, the challenge is securing sufficient computing resources without creating obligations that become difficult to support.
The outcome could affect not only Anthropic but also the wider market financing AI data centers and semiconductor equipment.
Investors Are Already Questioning a Trillion-Dollar Valuation
The scale of Anthropic’s prospective valuation has attracted considerable attention.
WWC ONE MEDIA previously reported on the debate in Anthropic Emerges as AI Market’s Last Major IPO Hope for 2026 — But a $1.5-Trillion Valuation and $80.7-Billion Loss Raise a Bigger Warning.
A valuation of $1.5 trillion would be extraordinary for a company founded in 2021.
Such a valuation would reflect expectations of substantial future revenue, profitability and competitive strength.
But valuations cannot be assessed through growth alone.
Investors must also consider cash flow, infrastructure obligations, customer concentration and the possibility that cheaper AI systems will reduce prices.
The IPO prospectus makes those trade-offs more visible.
And that may change the way Wall Street evaluates the entire AI industry.
Claude’s Enterprise Success Is the Biggest Argument for Anthropic
Anthropic’s strongest commercial advantage is its expanding enterprise business.
Claude has gained traction among developers and companies using AI for coding, analysis, research and workflow automation.
Businesses increasingly deploy advanced models through APIs, cloud marketplaces and enterprise products.
The resulting revenue can grow as more users adopt AI and existing customers increase usage.
But consumption-based pricing also introduces uncertainty.
Customers may reduce spending during economic downturns.
They may shift workloads to cheaper models.
And businesses can compare providers as AI tools become more interchangeable.
Anthropic therefore needs to keep improving its products while preventing operating expenses from rising too quickly.
A premium AI model can command strong prices today without guaranteeing the same pricing power in the future.
Competition From OpenAI, Google and Lower-Cost Models Remains Intense
Anthropic is competing against well-funded technology companies and emerging model developers.
OpenAI continues developing its own frontier models and commercial products.
Google operates a major AI research business while also controlling cloud infrastructure and widely used consumer platforms.
Other companies offer open-weight or lower-cost models that businesses can deploy through different service providers.
Competition may benefit customers through better performance and lower prices.
But it can make it harder for AI developers to maintain high margins.
A company that lowers prices to win customers must either reduce costs or accept lower profitability.
Anthropic’s challenge is especially difficult because competitors are investing heavily in the same underlying technologies.
Its ability to maintain a technological advantage will influence whether customers continue paying a premium for Claude.
Anthropic’s Founders Will Retain Extraordinary Voting Power
The IPO raises important governance questions as well.
Anthropic’s proposed ownership structure would give its seven co-founders control over approximately 50.1% of shareholder voting power through a special voting arrangement.
The company also operates as a public benefit corporation.
Its Long-Term Benefit Trust has an important role in the governance system.
These arrangements are designed partly to preserve Anthropic’s mission of developing advanced AI responsibly.
But they also mean ordinary public investors may have limited influence over certain company decisions.
The structure could become controversial if management chooses safety-related restrictions that reduce commercial opportunities or slow product development.
Investors must decide whether they accept that governance model before purchasing shares.
A Public Benefit Fund Adds Another Layer to the IPO Debate
Anthropic has also announced plans to dedicate a portion of its equity to public-benefit initiatives.
WWC ONE MEDIA covered the proposal in Anthropic Commits 5% of Its Shares to New Public Benefit Fund.
The plan is intended to support projects addressing some of the social consequences of advanced artificial intelligence.
It demonstrates how Anthropic is attempting to combine commercial growth with a broader public-benefit mission.
But there are financial and governance questions.
Investors may want to understand how the fund affects shareholder economics, who determines its priorities and how its activities will be overseen.
The allocation also should not be described as an immediate cash donation equivalent to 5% of a future market valuation.
The actual financial implications depend on implementation and the applicable arrangements.
AI Safety Could Be a Major Risk for Public Investors
Anthropic’s prospectus devotes extensive attention to the risks of increasingly capable AI.
These include misuse by malicious actors, cybersecurity failures, unpredictable model behavior, legal disputes and potential harms from advanced autonomous systems.
The company has publicly supported stronger safeguards and evaluation procedures.
WWC ONE MEDIA previously examined the industry response in OpenAI and Anthropic Turn to Independent AI Safety Watchdogs as Dangerous Model Behaviors Raise Alarm — But Who Will Police the Gatekeepers?.
The article discussed independent organizations evaluating advanced AI models and testing for dangerous capabilities.
These systems are important because increasingly powerful models may create risks beyond ordinary software defects.
But safety measures also create commercial and operational challenges.
Companies need to decide how to balance rapid deployment, customer demand, model capabilities and the resources devoted to testing.
For public investors, these decisions could become financially significant.
Regulators Are Watching the AI Industry
Anthropic is also operating in an environment of growing regulatory scrutiny.
WWC ONE MEDIA examined one aspect of that pressure in FTC Investigates OpenAI, Anthropic Over Potential Risks to Consumers.
The investigation concerns possible consumer-protection issues arising from advanced AI systems.
An investigation does not prove that a company violated the law.
Nevertheless, such inquiries can introduce compliance requirements, legal costs and uncertainty.
For an AI developer planning to become publicly traded, clear disclosure of regulatory exposure is particularly important.
Investors need to understand not only the company’s financial performance but also the legal and ethical risks associated with its technology.
What the Anthropic IPO Means for the Philippines and Asia
Anthropic’s prospective listing has implications for businesses throughout the Asia-Pacific region.
Companies are increasingly adopting AI for software development, customer support, research, document processing and operational automation.
The economics of leading model providers could influence how much businesses pay for these services.
If computing efficiency improves faster than infrastructure expenses grow, AI services may become cheaper or more capable.
But if cloud costs rise or providers face stronger financial pressure, enterprise customers could encounter higher prices or changes in service terms.
For Philippine businesses planning long-term AI investments, vendor reliability, pricing transparency, cybersecurity and data governance remain important considerations.
The Anthropic prospectus offers useful information about one leading supplier’s financial position.
It does not guarantee that Claude prices will rise or fall.
Nor does an IPO automatically improve the performance or safety of its products.
What Investors Still Need to Know
Anthropic has disclosed more information than ever before.
But several questions remain unanswered.
How much of the company’s future computing expenditure will support model training rather than existing customer demand?
Can its operating margin remain positive on an adjusted basis as the company develops more capable systems?
How will future cloud contracts affect cash flow?
Can Anthropic retain customers when competing models become cheaper?
Will its governance structure satisfy investors seeking influence over management decisions?
And what valuation will the public market ultimately assign to the company?
The final IPO prospectus, including updated offering terms and any additional disclosures, will be important.
Until then, speculation about the stock’s future performance should be treated cautiously.
The Bigger Question: Can AI’s Extraordinary Growth Pay for Itself?
Anthropic’s financial statements reveal the central paradox of the artificial intelligence boom.
Its products are generating enormous demand.
Revenue has increased at an extraordinary pace.
The economics of serving customers appear to be improving.
And the company is approaching operating profitability under conventional accounting measures.
Yet the cost of developing future AI models remains substantial.
Long-term infrastructure obligations could limit flexibility if growth slows.
Cloud providers, semiconductor companies, lenders and investors are all becoming financially connected to the continued expansion of AI demand.
That creates opportunities, but it also means a disappointment at one major company could have consequences beyond its own balance sheet.
The Bottom Line
Anthropic publicly filed its S-1 registration statement on October 26, 2026, setting the stage for a potential Nasdaq IPO under the symbol ANT.
The Claude developer reported $16.29 billion in first-half revenue, compared with $1.16 billion a year earlier.
Its reported net loss reached approximately $80.75 billion, largely because of noncash financing-related accounting charges.
Its operating loss narrowed to $693 million, while adjusted operating income turned positive at approximately $1.62 billion.
An implied gross margin of 85.4% suggests that serving customers can be highly profitable, but it excludes substantial costs involved in developing advanced AI models.
The company also faces enormous computing commitments, intense competition and questions about its future governance.
As Anthropic prepares for a potentially historic Wall Street debut, the defining question is no longer whether Claude can generate extraordinary revenue. It is whether Anthropic can convert that growth into lasting profits before the enormous cost of building the next generation of artificial intelligence catches up with it.
