Anthropic IPO: $60B Raise, $380B Valuation, Timeline & How to Invest

Anthropic doubled its annualized revenue from $9 billion to $19 billion in under four months. Read that sentence again. No enterprise software company, no cloud platform, no AI startup in history has ever scaled that fast at that magnitude. The company behind Claude is now in active discussions with Goldman Sachs and JPMorgan Chase about what could be the second-largest AI IPO in history: a $60 billion+ raise targeting October 2026, at a valuation that bankers are privately estimating between $400 billion and $500 billion.

This is the company that OpenAI’s own co-founders built after walking away from OpenAI. Dario and Daniela Amodei left because they believed the world’s most powerful AI systems needed a fundamentally different approach to safety and alignment. Three years later, eight of the Fortune 10 are Claude customers, and Anthropic’s enterprise revenue mix is stronger than any AI company preparing to go public.

But $19 billion in annualized revenue does not mean $19 billion in profit. Anthropic is burning cash at an extraordinary rate, with $12 billion earmarked for model training and another $7 billion for inference infrastructure in 2026 alone. Gross margins reportedly dropped to approximately 40% after inference costs surged 23% beyond internal projections, according to investor materials reviewed by The Information. For every dollar of revenue Anthropic generates, more than half disappears into GPU clusters before a single employee gets paid.

The S-1 filing, when it arrives, will force the market to reconcile the growth narrative with the burn reality. Here is everything investors need to know before Anthropic’s roadshow begins.

Last updated: April 3, 2026. Revenue and valuation figures sourced from Anthropic’s official announcements, The Information, CNBC, and Investing.com.

Anthropic IPO at a Glance

The Revenue Trajectory That Forced Wall Street to Pay Attention

The numbers tell a story that makes even the most aggressive SaaS growth curves look pedestrian. Anthropic crossed $19 billion in annualized revenue by March 2026. For context, that figure was $9 billion at year-end 2025, $1 billion in December 2024, and essentially zero in 2022. Revenue has grown more than 10x annually for three consecutive years, as TECHi previously covered in our analysis of Anthropic’s $183 billion valuation milestone.

Bloomberg first reported the $19 billion figure in early March, citing people familiar with the matter, around the time of a Morgan Stanley TMT conference where Amodei spoke. Much of the acceleration has been attributed to Claude Code, Anthropic’s agentic coding tool that has become a standard in enterprise software development. Claude Code alone generates over $2.5 billion in annualized revenue, with that number more than doubling since January 2026. Weekly active Claude Code users have also doubled in the same period.

The enterprise composition is what separates Anthropic from the pack. Approximately 80% of total revenue comes from business customers, making Anthropic’s revenue mix significantly more enterprise-weighted than OpenAI’s, which still derives a substantial portion from consumer ChatGPT subscriptions. Anthropic has over 300,000 business customers, with the number spending more than $100,000 annually growing 7x in the past year. Eight of the Fortune 10 are now Claude customers.

Anthropic projects revenue exceeding $70 billion by 2028, per TechCrunch reporting from November 2025. If even half that projection materializes, the current $380 billion valuation looks conservative relative to the growth rate. But revenue projections from pre-IPO companies are aspirational by design, and the path from $19 billion to $70 billion requires sustained enterprise adoption at a pace that has never been achieved in technology.

The Origin Story: When OpenAI’s Best Researchers Walked Away

Anthropic exists because of a philosophical schism at the most important AI laboratory in the world. In 2020 and 2021, a group of researchers left OpenAI, led by Dario Amodei (OpenAI’s Vice President of Research) and his sister Daniela Amodei (OpenAI’s Vice President of Safety and Policy). They took with them deep expertise in the scaling hypothesis, reinforcement learning from human feedback (RLHF), and a conviction that the industry’s most powerful models needed a different governance structure.

Dario had been instrumental in developing GPT-2 and GPT-3 at OpenAI. He understood, better than almost anyone outside the lab, that language models were going to become dramatically more capable with scale. His concern was not that they would fail; his concern was that they would succeed without adequate safety infrastructure. The departing group believed OpenAI’s rapid commercialization push, particularly the Microsoft partnership and ChatGPT launch, was outpacing its safety research.

Anthropic was incorporated in San Francisco in 2021 as a Public Benefit Corporation, a legal structure that obligates the company to consider societal impact alongside shareholder returns. The PBC designation is not merely symbolic; it gives Anthropic’s board legal standing to prioritize safety investments over short-term profitability, a distinction that could become material when public market shareholders start demanding faster path-to-profit timelines.

The founding team trained the first version of Claude in the summer of 2022 but chose not to release it publicly, citing the need for additional safety testing. Claude eventually launched in July 2023, nearly a year after ChatGPT’s debut. The delay was intentional: Anthropic bet that a safer, more reliable model would win enterprise trust even if it lost the consumer mindshare race to ChatGPT. That bet is paying off at $19 billion in annualized enterprise-heavy revenue.

Claude Code: The Product That Changed the Revenue Curve

Every great IPO story needs a product inflection point, and for Anthropic, that product is Claude Code. Launched as an agentic coding tool that can write, debug, refactor, and deploy software with minimal human oversight, Claude Code has become the fastest-growing product in Anthropic’s history.

The numbers: $2.5 billion in annualized revenue as of February 2026, with the run rate more than doubling since January 1. Business subscriptions to Claude Code have quadrupled in Q1 2026 alone. Enterprise use accounts for over half of all Claude Code revenue, meaning this is not a developer toy; it is becoming embedded in corporate software development workflows at Fortune 500 scale.

Claude Code’s traction matters for the IPO because it demonstrates a critical capability: Anthropic can build products, not just models. The AI industry is littered with companies that trained impressive foundation models but failed to build the application layer that generates recurring enterprise revenue. Anthropic has crossed that bridge. Claude Code, Claude for Enterprise, and the Claude API collectively form a product suite that enterprises are integrating into daily operations, creating the kind of workflow dependency that drives low churn and high expansion revenue.

The $30 Billion Series G and the Road to $380 Billion

Anthropic’s funding history reads like a venture capital fairy tale compressed into three years:

• 2023 (Series C): $450 million at $4.1 billion valuation. Led by Google and Spark Capital.
• 2023-2024 (Amazon Investment): Amazon committed $8 billion total across multiple tranches, establishing AWS as Anthropic’s primary cloud and training partner.
• March 2025 (Series E): $3.5 billion at $61.5 billion valuation. Led by Lightspeed Venture Partners, with Salesforce Ventures, Fidelity, and Jane Street.
• September 2025 (Series F): $13 billion at $183 billion valuation. The valuation tripled in six months.
• February 2026 (Series G): $30 billion at $380 billion valuation. Led by D.E. Shaw, MGX, and existing investors. Revenue had just crossed $14 billion annualized.

The investor roster now spans the full spectrum of institutional capital: Amazon ($8 billion total), Google ($3 billion), SoftBank, Andreessen Horowitz, Lightspeed, Fidelity, Salesforce Ventures, Menlo Ventures, Bessemer, and dozens of sovereign wealth funds and family offices. Microsoft provides access to Claude through its Azure Foundry platform, while Nvidia participated in later funding rounds, though specific investment amounts from both have not been independently confirmed. When every major cloud platform distributes Claude, it signals something beyond financial conviction: these cloud giants need Anthropic’s models on their platforms to compete.

The cloud platform distribution is a strategic asset few competitors can match. Claude is the only frontier AI model available on all three major cloud platforms: Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Azure Foundry. Enterprises already running workloads on AWS or Google Cloud can add Claude without separate procurement cycles, reducing friction and accelerating adoption.

IPO Timeline and What the S-1 Must Answer

According to The Information, Anthropic executives including CEO Dario Amodei have held internal discussions about launching an IPO as soon as October 2026. The company has engaged Wilson Sonsini as legal counsel and a banking consortium led by Goldman Sachs and JPMorgan to finalize the S-1 filing. Bankers expect the raise to exceed $60 billion, which would make it the second-largest technology IPO in history behind SpaceX.

The practical timeline from here: the S-1 filing would likely arrive in late summer 2026, followed by a 2-3 week roadshow, with pricing and listing targeting October. But several factors could accelerate or delay the schedule.

The competitive dimension is significant. OpenAI is also targeting a late 2026 listing, and Bloomberg has reported that both companies are racing to list first to capture institutional allocation budgets. If OpenAI files before Anthropic, it could absorb significant pent-up demand for AI exposure, dampening Anthropic’s reception. The reverse is also true. The sequencing of these two IPOs will shape how much capital each raises.

Critical questions the S-1 must answer for investors:

• Revenue recognition: The SEC may require Anthropic to change how it reports cloud computing credits as revenue. Amazon, Google, and Microsoft provide substantial compute credits as part of their investment agreements. If those credits are currently counted as revenue, the S-1 could show a materially different top-line figure than the $19 billion headline number.
• Gross margins: Inference costs reportedly surged 23% beyond projections in 2025, pushing gross margins to approximately 40%, according to investor materials. The S-1 will reveal whether margins are improving or deteriorating as Claude usage scales.
• Path to profitability: Anthropic projects positive free cash flow by 2027 and $17 billion in cash flow by 2028. The S-1 will detail the cost structure, capex commitments, and assumptions behind these projections.
• Customer concentration: What percentage of revenue comes from Amazon’s AWS reselling channel versus direct enterprise sales? High channel dependency would introduce concentration risk.

Anthropic vs. OpenAI: The IPO Showdown

The market will inevitably compare Anthropic and OpenAI as both prepare to list. Here is how the two stack up on the metrics that matter for IPO valuation:

OpenAI has larger headline revenue (approximately $29 billion in projected 2026 revenue vs. Anthropic’s $19 billion ARR) and a significantly larger consumer user base (900 million+ weekly active ChatGPT users). But Anthropic’s enterprise mix is stronger: 80% of revenue from business customers versus OpenAI’s more consumer-heavy composition. Enterprise revenue carries higher retention rates, better expansion economics, and lower churn.

OpenAI projects $14 billion in losses for 2026. Anthropic’s burn rate is lower in absolute terms ($12 billion in training costs plus $7 billion in inference), but the ratio of spending to revenue is comparable. Neither company is profitable. The critical difference: Anthropic projects positive free cash flow by 2027, while OpenAI has pushed its breakeven target to 2030.

From a governance perspective, Anthropic’s Public Benefit Corporation structure is a double-edged sword. It reassures safety-conscious investors and regulators, but it also gives the board legal authority to prioritize safety over shareholder returns. For growth-oriented public market investors accustomed to management teams maximizing earnings per share, the PBC structure may require an adjustment in expectations.

PitchBook’s analysis concluded that among the three major AI IPO candidates (OpenAI, Anthropic, and Databricks), Anthropic offers the strongest combination of growth rate and business quality fundamentals. That assessment, if validated by the S-1, could make Anthropic the preferred institutional allocation over OpenAI despite the smaller headline revenue.

Risk Factors Every Investor Must Weigh

Cash Burn Is Structural, Not Temporary

Anthropic plans to spend approximately $12 billion training models and $7 billion on inference infrastructure in 2026. That $19 billion in planned spending roughly equals the company’s annualized revenue, meaning Anthropic is spending every dollar it earns and then some. The company burned approximately $2.8 billion in cash in 2025, and the 2026 figure will be substantially higher. Without the IPO proceeds, Anthropic would need another massive private raise within 12-18 months.

Gross Margin Pressure From Inference Costs

Inference costs reportedly surged 23% more than expected in 2025, pushing gross margins to approximately 40%. As Claude usage scales across 300,000+ enterprise customers, the inference cost base grows proportionally. If Anthropic cannot reduce per-query costs faster than usage grows, margins will compress further. CEO Dario Amodei has acknowledged this risk publicly, stating that revenue growth may not keep pace with expanding operations.

Revenue Recognition and Cloud Credit Accounting

The SEC is expected to scrutinize how Anthropic accounts for compute credits from Amazon, Google, and Microsoft. If a significant portion of the $19 billion ARR reflects cloud credits rather than cash revenue, the S-1 could present a materially different financial picture than the pre-IPO narrative suggests. This accounting question is one of the most important open issues for both Anthropic and OpenAI ahead of their respective filings.

Competitive Intensity in Frontier AI

Anthropic competes against OpenAI, Google DeepMind, Meta’s Llama (open-source), and an emerging wave of Chinese AI labs. The frontier AI market is not a winner-take-all market; it is a capability race where any competitor that achieves a meaningful performance leap can rapidly capture market share. Anthropic’s position is strong today, but the technology landscape shifts in months, not years.

Regulatory Uncertainty

AI regulation is evolving rapidly across the U.S., EU, and other major markets. Anthropic’s safety-first positioning may become a competitive advantage if regulation tightens, but mandatory compliance costs could also compress margins. The company’s PBC structure adds a layer of complexity: if safety mandates conflict with revenue optimization, Anthropic’s board has a legal obligation to prioritize safety. Public market shareholders may not always appreciate that calculus.

How to Invest in Anthropic Stock

Before the IPO

Anthropic shares are not available through standard brokerage accounts. Pre-IPO access is restricted to accredited investors through secondary market platforms (EquityZen, Forge Global, Hiive) and select institutional channels. The KraneShares Artificial Intelligence and Technology ETF (AGIX) is currently the only publicly listed ETF with direct pre-IPO ownership exposure to Anthropic.

Indirect Exposure Through Existing Public Stocks

Investors who want Anthropic exposure before the IPO can access it through its largest corporate investors. Amazon (AMZN) holds $8 billion in Anthropic equity and resells Claude through AWS Bedrock. Google/Alphabet (GOOGL) holds $3 billion and distributes Claude through Vertex AI. Both companies have already realized significant paper gains on their Anthropic positions. For a broader view of AI investment opportunities, see our guide to the best AI stocks. This indirect exposure offers diversification that a concentrated pre-IPO position cannot.

After the IPO

Once Anthropic lists, shares will be available through any brokerage account. The lock-up period (typically 90-180 days) and the first earnings report after listing will be the key events that reveal whether the IPO price was warranted. Disciplined investors may find better entry points during lock-up expiry or post-earnings volatility rather than chasing the first-day pop.

The Bull Case and the Bear Case

IPO Timeline: Key Dates to Watch

• Q2 2026 (Current): S-1 preparation underway with Wilson Sonsini and Goldman Sachs/JPMorgan consortium. SEC pre-filing discussions on revenue recognition and cloud credit accounting.
• July-August 2026: Expected S-1 filing with SEC. First public look at audited financials, customer metrics, and segment disclosures.
• September 2026: IPO roadshow. Dario Amodei and Daniela Amodei present to institutional investors globally.
• October 2026 (Target): Nasdaq listing and first day of public trading.
• Q1 2027: Lock-up expiry. Pre-IPO investors and employees can sell shares for the first time.
• Q1 2027: First full-year earnings report as a public company. The market’s first comprehensive look at Anthropic’s financial trajectory.

The AI industry is entering its public market phase. SpaceX filed its confidential S-1 on April 1. Cerebras is targeting an April listing. OpenAI and Anthropic are both racing toward late 2026 debuts. Within 18 months, the companies building the infrastructure for artificial general intelligence will be publicly traded, subject to quarterly earnings scrutiny, and priced by the same market that values Nvidia at $4.3 trillion. Anthropic’s safety-first DNA, enterprise-heavy revenue, and $19 billion growth trajectory make it one of the most consequential IPOs of the decade. Whether the price is right depends entirely on what the S-1 reveals about the gap between revenue and reality.

Frequently Asked Questions