Anthropic targets US$1.4 trillion valuation in blockbuster IPO
The Claude chatbot maker and its co-founders are seeking to raise as much as US$90.4 billion
ANTHROPIC is seeking a fully diluted valuation of about US$1.4 trillion in its initial public offering, in a bid to make history as the first frontier artificial intelligence lab to go public.
The Claude chatbot maker and its co-founders are seeking to raise as much as US$90.4 billion in the IPO and a private placement. The base offering would raise up to US$64.4 billion, with shares marketed for US$70 to US$78 each, according to a filing on Tuesday (Nov 10) with the US Securities and Exchange Commission.
Anthropic is offering 801.4 million shares, while backers are selling 23.6 million shares.
The figure also includes a private placement set to raise US$26 billion in aggregate, to Advanced Micro Devices, Broadcom, Alphabet’s Google, Micron Technology, Microsoft, Nvidia and Taiwan Semiconductor Manufacturing Co, the filing showed.
At the top of the price range, Anthropic would have a market value of about US$1 trillion, based on the shares outstanding listed in the filing. Accounting for employee stock options, restricted shares and stock reserved for the company’s charitable donation matching programme, Anthropic would have a fully diluted value of about US$1.4 trillion.
Anthropic’s ambitious target, including the private placement, would hit roughly US$100 billion when including an over-allotment option or greenshoe, Bloomberg News calculations showed.
That would top SpaceX’s US$75 billion base offering, or the US$86.2 billion it raised with the greenshoe.
The goal defies jitters from a sharp sell-off in AI-related stocks last week, driven by expectations that politicians promising tougher regulation of data centres and related infrastructure will drive up costs and limit development of the technology.
Investors are under pressure to decide whether valuations in AI reflect a plausible trajectory for a technology that many expect to transform the economy and society as a whole, or overconfidence in an investment theme that has yet to prove its staying power.
The widespread adoption of Claude Code and rival tools by software developers point to the potential transformation of a wide range of professional activity, from discovering new miracle drugs to navigating complex legal and accounting rules to running a factory.
Already, more than 90 per cent of contributions to Anthropic’s code base was written by Claude, the filing showed.
Anthropic has introduced so-called vertical solutions in these areas, with particular emphasis on healthcare and life sciences. It counts Novartis CEO Vas Narasimhan as one of its board members.
The company’s seven co-founders, including CEO Dario Amodei, are offering 3.38 million shares each in the IPO, the filing showed.
The IPO is set to price on Nov 18 and trade the following day, according to terms of the deal reviewed by Bloomberg.
Sales are surging. The company reported a revenue run rate of US$81 billion as at Oct 31, up US$10 billion from the end of September, the filing showed. It had a US$65 billion run rate figure at the end of July, Bloomberg News reported.
Anthropic defines the metric as annualising consumption-based fees for a trailing 28-day period and subscription-based fees for a single month.
The company reported third-quarter revenue of nearly US$17 billion, versus US$1.4 billion in the same period a year earlier.
On Monday, Anthropic unveiled new versions of its powerful Mythos and Fable models, both of which are faster than prior versions, while Fable is less expensive, as some companies cut back on their spending on the technology.
Keeping the pace of development running at a breakneck pace will require growing revenue faster than Anthropic’s spending on AI infrastructure. Its commitments climbed to nearly US$901 billion in the latest filing, most of which are for computing power, chips.
About US$71 billion worth – a figure larger than the individual market caps of roughly two-thirds of S&P 500 companies – are due just in the year through Sep 30, 2027.
Unlike other tech companies with sources of income such as advertising to fall back on, Anthropic will have to convince investors that its focus on AI alone will generate profit far beyond the technology’s surging costs.
SpaceX may have tied its aspirations to AI when it pitched investors on its IPO, but most of its actual revenue at the time came from its other business lines.
That has not stopped asset managers, venture capital (VC) firms and some of the biggest tech companies from backing Anthropic to the tune of tens of billions of dollars.
Funds scattered across investment houses including Fidelity Investments, Capital Group and BlackRock and VCs such as Iconiq, Spark Capital and Sequoia Capital hold Anthropic shares before the IPO.
Google, Amazon and Microsoft are backers, as well as customers and key suppliers of computing power.
Pace the frontier
The speed of development has become increasingly controversial, with Anthropic’s CEO Dario Amodei among those calling for top AI labs to “pace the frontier” in order to prioritise safety.
The company itself warned that the technology could pose “catastrophic or existential risks to humanity” in its IPO filing, if the technology is not developed with adequate safeguards and oversight.
Regulators, investors and even Anthropic’s employees have flagged the significant risk of humans losing control of the technology, with one putting the chance of AI eliminating all humans in the next decade at greater than 10 per cent. Bill Gates has said AI is powerful enough to drive events that cause “a billion deaths”.
Anthropic’s rival OpenAI is targeting a 2027 listing; CEO Sam Altman called 2026 an “ill-advised moment” for an IPO, citing concern over AI safety. The company is raising funds at a US$1.25 trillion valuation, people familiar with the matter have said.
Anthropic’s origin story centres around former OpenAI executives defecting because they wanted to prioritise AI safety.
Nearly six years after its founding, Anthropic has structured itself as a public benefit corporation that balances shareholder interests with that of society, and whose governance includes a long-term benefit trust that has the power to elect and remove a majority of the members of the company’s board.
Anthropic’s seven co-founders hold roughly 1.4 per cent stakes in the company, making them each billionaires several times over. That remains true even if they stick to their pledges to give 80 per cent of their personal equity in the company to philanthropic causes.
The company’s total operating expenses jumped to US$17.5 billion in Q3, a 292 per cent increase from a year prior. It flipped from an operating profit of US$974 million in Q2 – its first after at least five quarters of operating losses – back to a negative US$567 million in the three months ended Sep 30, according to the filing.
The operating loss in the third quarter was primarily driven by US$5 billion of computing costs under short-term commitments and US$2 billion of non-cash charges related to the company’s charitable donation-matching programme.
AI pricing pressures have begun to hit Anthropic’s revenue. Lower prices on some of its new models, plus contractual discounts for some customers, reduced its revenue growth in Q3 by about US$600 million, the filing shows.
Anthropic has cut prices in newer model roll-outs, including Opus 5.5 in September; OpenAI has also introduced lower-priced models in recent months.
The AI giants face growing competition from cheaper open-weight models, including from Chinese developers, which customers can download and fine-tune to their needs.
Goldman Sachs and Morgan Stanley are leading Anthropic’s IPO, with participation from 20 other banks, the filing showed. BLOOMBERG
Decoding Asia newsletter: your guide to navigating Asia in a new global order. Sign up here to get Decoding Asia newsletter. Delivered to your inbox. Free.
Share with us your feedback on BT's products and services
TRENDING NOW
No Malaysian probe yet into Nvidia customer Megaspeed’s unit; firm earlier cited ministry checks
Nvidia, CEO Jensen Huang sued over disclosures related to suspected Chinese spy
About 3 million Singaporeans to receive S$100 to S$250 Assurance Package cash payouts in December
Chery recalls about 54,000 vehicles in China after Malaysia axle-failure incidents