Anthropic Closes Its First Day of Trading at a $1.7 Trillion Valuation, Raising $100 Billion
Anthropic went public in the United States on November 19 under the ticker ANT, closing its first day with a market capitalization exceeding $1.7 trillion and raising approximately $100 billion through a public offering, private placements, and other means. The company’s annualized revenue was about $81 billion at the end of October 2026, but its net loss for the first three quarters was approximately $104 billion, while its future computing power procurement and leasing commitments totaled more than $900 billion.
Why read this
SelectedAI score 95
The article reviews Anthropic’s IPO financing, shareholder returns, and computing power commitments, highlighting the revenue growth and operational pressures behind its trillion-dollar valuation.
Topics
Body · AI translation
The largest IPO in history has arrived.
On November 19, Anthropic went public in the United States under the ticker symbol ANT. On a fully diluted basis, the company’s market capitalization briefly surpassed $2 trillion and exceeded $1.7 trillion at the close of its first trading day—more than RMB 11 trillion. Through a public offering, private placements and other means, Anthropic raised as much as $100 billion in total, overtaking SpaceX as the largest IPO in history.
Led by Dario Amodei and Daniela Amodei, siblings born in the 1980s, they and a group of former OpenAI colleagues took just over five years to build Anthropic from scratch into a global giant, creating extraordinary wealth in the process—the stakes held by each of its seven co-founders are worth more than $10 billion.
At the same time, a group of employees is also set to achieve financial freedom.
The prospectus shows that the company still has 1.774 billion unexercised stock options and another 1.07 billion unvested restricted stock units, or RSUs. Based on the closing price of $97.17, the potential value of these equity awards exceeds $270 billion. Anthropic currently has about 5,300 employees. A simple average based on that headcount would amount to more than RMB 300 million in equity per employee.
Investors have also reaped substantial rewards. Anthropic’s shareholders range from early investors such as former Google chairman and CEO Eric Schmidt to global giants including Google, Amazon and Nvidia; investment firms such as Sequoia Capital, Spark Capital and DST Global; and sovereign wealth funds including GIC and the Qatar Investment Authority.
As the IPO bell rang, they all shared in this vast wealth bonanza.
Former OpenAI Employees Built a Trillion-Dollar IPO
Anthropic may be OpenAI’s leading rival, but it was founded by a group of former OpenAI employees.
In 2021, Dario Amodei and his sister Daniela Amodei co-founded Anthropic with former OpenAI colleagues Tom Brown, Jack Clark, Jared Kaplan, Sam McCandlish and Christopher Olah, among others.
Their dissatisfaction with OpenAI’s accelerated commercialization and concern that AI safety was being neglected prompted their collective departure. In the founding team’s view, as artificial intelligence became increasingly capable, greater attention needed to be paid to safety research and technological governance.
That decision was closely tied to the siblings’ own experiences.
Dario Amodei studied at institutions including Stanford, Caltech and Princeton, and worked at companies including Baidu and Google. In 2016, he joined OpenAI to conduct AI safety research and participated in the development of cutting-edge models including GPT-2 and GPT-3.
Following in her brother’s footsteps, Daniela Amodei, who had studied English literature at the University of California, Santa Cruz, joined OpenAI in 2018 and ultimately became the company’s vice president for safety and policy.
One understood models and the other governance, a complementary combination that defined Anthropic from the outset. At its founding, Anthropic described itself as an AI safety and research company whose goal was to build AI systems that were “reliable, interpretable and controllable.”
In 2023, Anthropic officially launched Claude, beginning to turn its previous model research capabilities into products and services for the market. As the models continued to improve, Anthropic’s business gradually expanded beyond general-purpose AI models into specific use cases such as enterprise development, software programming and office work.
Among these offerings, Claude Code became an important product in Anthropic’s commercialization and is described by the company as a “virtual collaborator” capable of planning, writing, testing and debugging code. Unlike tools that merely generate code through conversation, Claude Code can read codebases, modify files, execute commands and continuously carry out a series of tasks including planning, writing, testing and debugging.
This change in product capabilities directly affected Anthropic’s business model. As the chains of tasks Claude Code could perform grew longer, enterprises gradually shifted from using AI for one-off questions and answers to incorporating it into continuous workflows. In the process, models were called repeatedly, and Token consumption increased as tasks progressed.
For Anthropic, this meant that enterprise customers’ spending was no longer limited to fixed subscription fees but could also grow with actual usage. In the first half of 2026, about 80% of the company’s revenue came from usage-based charges, totaling $13.1 billion, while the remaining approximately 20% came from subscriptions to products such as Claude.
Market acceptance also put its annualized revenue on a steep growth curve.
According to Dario Amodei, Anthropic’s revenue has grown by roughly tenfold every year since it generated its first revenue. From about $10 million in 2022 to around $1 billion in December 2024, the figure had reached approximately $81 billion by the end of October 2026.
One reason behind this surge in performance is that more and more companies have begun using Claude in real-world business operations while scaling up their model usage. Anthropic itself is also a major user of Claude Code—in September 2026, Claude wrote more than 90% of the code the company committed to its internal repositories.
To capture a larger share of the market, Anthropic has also continued expanding its product portfolio. The company now has model families including Fable, Mythos, Opus, Sonnet, and Haiku. As its IPO approached, it launched products such as Fable 5.2 and Mythos 5.2 in rapid succession, while extending its AI capabilities further into non-programming office work.
From focusing on AI safety research when it was founded, to launching Claude in 2023, and then achieving explosive revenue growth through various agents, Anthropic took just over five years to transform from a technology research organization into a commercial AI company and ultimately grow into a global giant valued at more than $1 trillion.
A Wealth Bonanza Begins, Delivering Huge Gains for Investors
Backed by strong investor confidence, Anthropic raised funding all the way through a Series H round.
Shortly after it was founded in 2021, Skype co-founder Jaan Tallinn, Facebook co-founder Dustin Moskovitz, and former Google chairman and CEO Eric Schmidt invested $124 million in the startup team that had left OpenAI.
At the time, Anthropic had yet to launch Claude and did not have the vast enterprise customer base it later developed. What investors valued more was the founding team’s experience in cutting-edge artificial intelligence research and its judgment about the direction of AI technology.
This early funding provided crucial support for Anthropic to build its team and develop models. But as competition in large models intensified, Anthropic soon needed more capital.
In 2023, Google became the first to invest $300 million in Anthropic. The company later completed a $450 million Series C round with participation from institutions including Spark Capital. In September of the same year, Amazon announced a $4 billion investment in the company, after which Google decided to invest an additional $2 billion.
Google and Amazon were not only providers of capital but also major cloud-computing service providers capable of supporting Anthropic’s model training and deployment through their cloud infrastructure. This partnership combining capital and computing power created important conditions for Anthropic’s subsequent expansion.
After Claude entered the commercialization stage, the company accelerated its fundraising further.
In March 2025, Anthropic completed a $3.5 billion financing round backed by investors including Lightspeed, Bessemer Venture Partners, Cisco Investments, D1 Capital, Fidelity Management & Research, General Catalyst, Jane Street, and Salesforce Ventures. Six months later, it raised another $13 billion from investors including Lightspeed, Blackstone, Goldman Sachs, and the Qatar Investment Authority.
By 2026, the company had completed two major financing rounds in the first half of the year, sending its valuation soaring from $380 billion to $965 billion. Investors in its $30 billion Series G round included GIC, Coatue, Google, Amazon, Sequoia Capital, Temasek, Microsoft, and Nvidia.
Its $65 billion Series H round was backed by Altimeter Capital, Dragoneer, Greenoaks, Sequoia Capital, Capital Group, Coatue, D1 Capital, GIC, Blackstone, DST Global, MGX, Lightspeed, Amazon, Micron, Samsung, and SK hynix, among others.
As its listing plans advanced, the company began seeking larger amounts of capital from the public markets. In addition to its public share offering, Anthropic will conduct a $26 billion private placement with companies including Google, Nvidia, Microsoft, Broadcom, Micron, TSMC, and AMD.
After the IPO and concurrent private placement are completed, Amazon and Google are expected to hold nearly 15% and 14% of Anthropic, respectively, making them major shareholders in the company. After years of sustained investment, the two technology giants ultimately earned substantial returns by providing Anthropic with funding and computing-power support.
Other investors also reaped handsome rewards. Returns varied enormously depending on whether they invested when Anthropic was valued at only a few billion dollars in its early days or at $965 billion six months ago. Yet with the company now worth $1.7 trillion, even investors who entered in the final pre-IPO round have made considerable gains.
As the opening bell rang, the seven co-founders also cashed out by selling some of their shares. According to media reports, they will each sell approximately 3.3784 million shares, for a combined total of about 23.65 million shares and proceeds exceeding RMB 10 billion.
Although their equity stakes will decline, control of the company will remain in the hands of the founding team. Under Anthropic’s plan, in addition to retaining its status as a public benefit corporation, the seven co-founders will secure approximately 50.1% of the voting power through Class F shares held by Founder LLC.
In the AI Era, IPOs Cross the Trillion-Dollar Threshold
As a new wave of artificial intelligence sweeps across the globe, trillion-dollar IPOs are emerging one after another.
SpaceX and Anthropic went public less than six months apart, yet both surpassed $1 trillion in market capitalization. Market reports suggest that OpenAI, which has already embarked on the path toward an IPO, is raising a new round of funding at a pre-money valuation of $1.25 trillion and is expected to go public in 2027.
The stories behind these companies’ listings share one common theme: AGI can create tremendous value.
SpaceX stated bluntly in its prospectus that it has already secured the largest serviceable market in human history—a potential total value of up to $28.5 trillion, of which AI accounts for $26.5 trillion.
Anthropic, meanwhile, paints a picture in which knowledge work generated more than $40 trillion in economic value worldwide in 2025, with its opportunity lying in using AI to boost productivity and accelerate economic growth.
It is precisely these enormous expectations for potential markets that have made investors willing to assign leading AI companies valuations far above those of traditional businesses. However, as concerns about an AI bubble continue to mount and the AI industry enters a phase of large-scale commercialization, capital markets have begun to reassess these long-term expectations.
Since filing its prospectus, Anthropic has also faced disagreement over its valuation.
The market had previously speculated that the company could achieve a valuation of $2 trillion. Following changes in the external environment, coupled with signs of slowing performance, valuation expectations for its IPO have also declined.
In the third quarter of 2026, it generated approximately $16.96 billion in revenue, up more than 1,100% year over year. On a sequential basis, however, revenue grew by about 47% from the second quarter, markedly below the roughly 144% growth recorded in the second quarter.
At the same time, Anthropic’s profitability also fluctuated. The company posted operating profit of approximately $974 million in the second quarter of 2026, but swung to an operating loss of about $567 million in the third quarter. In addition, due to factors including changes in the fair value of convertible notes and derivative liabilities, the company recorded a net loss of approximately $104 billion in the first three quarters of 2026.
More noteworthy than the accounting loss is Anthropic’s continually rising spending on computing power.
As more companies integrate Claude into their operations and continuously run coding agents and other AI tools, Anthropic must provide more inference computing capacity. To keep its models competitive, it must also continue developing and training next-generation models, requiring sustained investment in vast amounts of training resources.
As the business continues to expand, demand for computing power will persist. According to Anthropic, its future commitments to purchase and lease computing capacity from partners including Amazon and Google have already exceeded $900 billion in total.
Nevertheless, the massive investment in computing power and pressure on profitability have not dampened investor enthusiasm.
According to media reports, institutional investor demand for the company’s shares totaled approximately $400 billion, while retail demand approached $100 billion. This prompted Anthropic to raise its offer price from a range of $70 to $78 per share to more than $80, while also increasing the number of shares offered.
For Anthropic, reaching $8.1 billion in annualized revenue within five years merely demonstrates the enormous commercial potential of frontier AI technology. Going forward, controlling computing costs and improving operating performance while maintaining the competitiveness of its models will be crucial to supporting its trillion-dollar market capitalization.
This article is from the WeChat public account “ChinaVenture,” written by Lu Zhigao and edited by Wang Qingwu. It is republished by 36Kr with permission.
AI Trends