AI startup Anthropic is preparing for an unprecedented IPO, planning to raise up to $100 billion at a valuation of up to $2 trillion, potentially surpassing SpaceX as the largest public listing in history. Sources familiar with the matter say Nvidia is in talks to serve as a cornerstone investor with a subscription of up to $10 billion. Anthropic’s annualized revenue surpassed $65 billion as of the end of July, up sharply from $9 billion at the end of 2025, with the company projecting 2028 revenue of $190 billion to $200 billion. The company maintains close ties with Nvidia while also diversifying its compute sources across Amazon AWS, Google TPUs, and Broadcom. The IPO is expected to be completed before the U.S. midterm elections in November, serving as a critical test of the public market’s appetite for sky-high valuations of top-tier AI companies.

AI startup Anthropic is preparing for an initial public offering (IPO) that could rewrite records across global capital markets. Sources familiar with the matter say the five-year-old company plans to raise up to $100 billion (approximately NT$3.2 trillion) through the listing, giving it a valuation of up to $2 trillion (approximately NT$63.3 trillion). Even more notably, Anthropic is in talks with chip giant Nvidia (NVDA) for the latter to serve as a cornerstone investor with a subscription of up to $10 billion (approximately NT$320 billion).

Two sources familiar with the matter disclosed the plan, which remains under negotiation. One said Nvidia is considering investing up to $10 billion in Anthropic’s IPO, but discussions are still at a preliminary stage and terms could change at any time. Anthropic declined to comment, while Nvidia has not yet responded to requests for comment.

If the deal goes through as planned, it would match or even surpass the IPO record set by SpaceX in June of this year, becoming the largest public listing in history. For Anthropic, locking in a deep-pocketed cornerstone investor like Nvidia before formally marketing shares to the market amounts to a vote of confidence in this massive fundraising, helping to boost participation appetite among other institutional investors.

The Deep Ties Between Nvidia and Anthropic

Nvidia’s entry as a cornerstone investor would further deepen what is already an extremely close commercial relationship. Anthropic’s core AI computing currently relies heavily on Nvidia graphics processing units (GPUs), and Nvidia announced as early as November last year that it would invest up to $10 billion in Anthropic under a broader collaboration framework. Under that agreement, Anthropic committed to purchasing $30 billion worth of Microsoft (MSFT) Azure computing capacity powered by Nvidia chips, with Microsoft itself also investing up to $5 billion.

However, Anthropic has been actively diversifying its chip supply sources in recent years to avoid over-reliance on any single supplier. Amazon (AMZN) and Google parent Alphabet (GOOGL) are both shareholders in Anthropic and its primary compute providers. In April of this year, Anthropic committed to spending more than $100 billion on Amazon Web Services (AWS) over the next decade, using more than 1 million of Amazon’s in-house Trainium2 chips, while Amazon announced an additional investment of up to $25 billion.

Anthropic has also reached agreements with Google and Broadcom (AVGO) to add several gigawatts (GW) of TPU (Tensor Processing Unit) computing capacity. As demand for its Claude chatbot has surged and available compute has become increasingly constrained, the company has even formed an internal team to design custom chips tailored specifically for Claude, aiming to gain greater control over hardware costs.

A defining feature of this supply chain is the heavy overlap between Anthropic’s shareholders and its compute suppliers — the same group of companies both invests in the company and sells it computing power, creating a two-way flow of capital and compute:

Valuation and Revenue Surge

Anthropic’s valuation growth has been rare even by tech industry standards. The company raised $65 billion (approximately NT$2.1 trillion) in May of this year at a valuation of $965 billion (approximately NT$30.5 trillion). Now the IPO target valuation has reached up to $2 trillion — more than doubling in just four months. Zooming out, the trajectory is even steeper: the Series F round in September 2025 valued the company at $183 billion, and by the Series G round in February 2026, that figure had reached $380 billion.

Revenue performance has been equally staggering. As of the end of July this year, Anthropic’s annualized revenue had surpassed $65 billion, far above the roughly $9 billion (approximately NT$280 billion) level at the end of 2025; the figure had only just crossed $30 billion in April of this year. The company projects 2028 revenue of $190 billion to $200 billion (approximately NT$6 trillion to NT$6.3 trillion), a forecast that underpins its valuation expectations.

Key recent data points for Anthropic:

Metric	Figure
September 2025 Series F valuation	$183 billion
February 2026 Series G valuation	$380 billion
May 2026 Series H valuation	$965 billion
IPO target valuation	approximately $2 trillion
IPO target raise	Up to $100 billion
End-2025 annualized revenue	approximately $9 billion
April 2026 annualized revenue	Surpassed $30 billion
End-July 2026 annualized revenue	Surpassed $65 billion
2028 revenue projection	$190 billion–$200 billion

Note: Figures above are compiled from multiple reports and are denominated in U.S. dollars.

Japan Expansion: Tokyo Office and NEC Partnership

Anthropic’s expansion is not limited to the United States. In October 2025, the company opened its first Asia-Pacific office in Tokyo and signed a memorandum of understanding with Japan’s AI Safety Institute (AISI) for research and evaluation collaboration.

In April 2026, NEC announced a strategic partnership with Anthropic, becoming the first Anthropic global partner in Japan. Approximately 30,000 NEC Group employees will adopt Claude, and the two companies will jointly develop secure, industry-specific AI solutions for sectors including finance, manufacturing, and local government. NEC’s cybersecurity monitoring services have also adopted Anthropic’s technology. NEC noted in its statement that Japanese enterprises commonly face barriers to AI adoption including IT talent shortages, difficulty accumulating operational experience, and stringent security and Japan-specific regulatory compliance requirements — with particularly high demands for safety and transparency in the financial and public administration sectors.

Market Test and Timeline

Anthropic expects to complete its listing before the U.S. midterm elections in November. This unprecedented transaction will serve as a major litmus test for whether public markets can accept the sky-high valuations and massive capital requirements of top-tier AI companies.

The U.S. IPO market has already been exceptionally hot this year. According to Dealogic, excluding special purpose acquisition companies (SPACs), cumulative U.S. IPO proceeds through the end of August reached a record $137 billion (approximately NT$4.3 trillion). Against this market backdrop, a successful Anthropic mega-IPO would set an important benchmark for subsequent AI startup listings.

It is worth noting that cornerstone investors are playing an increasingly significant role in ultra-large-scale IPOs. When chip designer Arm went public, cornerstone investors included Nvidia and Amazon; SpaceX’s earlier listing brought in Saudi Arabia’s Public Investment Fund (PIF) as one of its cornerstone investors. For Anthropic, Nvidia’s participation represents not only financial backing but also a powerful signal of endorsement from the AI chip supply chain for its business model.

https://finance.biggo.com/news/40801a4c-ccb5-439a-b622-a85ff3034370?shem=aimgspc,