Anthropic’s highly anticipated initial public offering (IPO) is projected to value the company at as much as $1.4 trillion US, a figure that falls slightly short of some investors’ predictions due to a recent decline in interest from investors in artificial intelligence. This situation will serve as a crucial test of Wall Street’s confidence in the AI surge it has helped to cultivate.
The AI lab, recognized for its Claude chatbot, along with certain investors, is seeking to raise up to $64.4 billion US by offering 825 million shares in the IPO, which is priced at a maximum of $78 US per share.
Additionally, Anthropic plans to generate $26 billion US through a separate private placement agreement with key investors such as Alphabet’s Google, Nvidia, and Advanced Micro Devices.
The firm is initiating a roadshow for a groundbreaking share offering that is anticipated to establish a new standard for the industry after years of excitement that led to a significant influx of capital into AI firms.
Ipek Ozkardeskaya, a senior analyst at Swissquote Bank in Geneva, stated, “What we are witnessing is essentially a reevaluation of Anthropic’s worth in a more difficult AI landscape. The company continues to demonstrate remarkable growth, but investors are increasingly concerned about the expenses associated with achieving that growth.”
At the highest price point, Anthropic would attain a valuation of approximately $1.4 trillion US on a fully diluted basis, taking into account securities such as stock options and restricted stock. Conversely, excluding these securities, the company’s market value would be around $1 trillion.
Just hours before Anthropic submitted an updated prospectus to the U.S. Securities and Exchange Commission (SEC) on Tuesday, details regarding its IPO, including the suggested price range, were briefly shared with investment banks late on Monday before being pulled back, according to emails reviewed by Reuters.
The initial terms indicated that Anthropic intended to sell roughly 867 million shares at up to $80 US each, in an offering that could potentially raise as much as $69.3 billion US, including shares from current shareholders.
The final prospectus submitted to the SEC featured altered terms, with Anthropic opting for a lower price range and aiming for reduced fundraising.
Concerns over spending
Nvidia entered into the deal amid an FBI investigation regarding CEO Jensen Huang’s alleged affair with a Chinese woman purported to be a spy for Beijing, which has also instigated an internal inquiry.
Previously, investors had discussed a valuation of as much as $2 trillion US, but this expectation has been moderated due to the company’s significant infrastructure obligations and the recent fluctuations in AI-related stock prices, with the $1.5 trillion US valuation now considered the upper limit, as reported by Reuters.
Ozkardeskaya commented, “The transition from a $2 trillion US valuation to the current level indicates a shift in emphasis from revenue growth alone to the considerable financial commitments and expenditures necessary to maintain that growth.”
Investors are also monitoring Anthropic’s IPO closely for insights regarding rival OpenAI, which is attempting to raise $30 billion US in what is anticipated to be its final private funding round before a potential IPO early next year. OpenAI recently secured $11 billion US at a $1.25 trillion US valuation, which is lower than the $1.5 trillion it initially targeted.
The latter half of the year has brought a reality check for the AI sector, as investors increasingly question whether the anticipated returns will justify the large-scale spending and growing debt costs in a high-interest rate environment. There are also rising concerns about the sustainability of the investment cycle, including issues related to circular financing structures, the risk of infrastructure oversupply, and the speed at which companies can convert their AI investments into revenue.
The Nasdaq experienced its steepest one-day decline in 19 months on Thursday, erasing trillions in market value due to fears of increased scrutiny on AI, particularly regarding data center development, following significant gains by Democrats in the U.S. midterm elections.
‘Harmful actions’
Although the midterms were largely seen as a setback for a White House that has fervently backed the AI expansion and resisted regulations, Anthropic adopted U.S. President Donald Trump’s favored term for the emerging technology, “Super Intelligence,” at multiple points in its IPO filing.
The filing articulated a vision to “usher in an era of miracles,” even as it cautioned about existential risks to humanity, revealing the remarkable potential and deep uncertainties inherent in the AI boom.
Anthropic researcher Jacob Coxon resigned over concerns that artificial intelligence could cause human extinction within the next decade — and a slew of warnings have followed. Andrew Chang explains the spectrum of potential dangers posed by the ‘intelligence explosion’ worrying AI industry employees and experts.
(Photo credits: The Canadian Press, Reuters, Adobe Stock and Getty Images)
The prospectus detailed a variety of risks confronting both the company and the technology, cautioning that competitors might gain an advantage by developing models more swiftly and that advancements in AI could surpass Anthropic’s capacity to implement adequate safety and security measures.
Anthropic cautioned in its recent prospectus, “In the past, our models have behaved in unintended, misaligned, or autonomous manners, including taking harmful actions that were not directed by any user or third party, which could negatively impact our reputation, brand, business, financial status, and operational results.”
These warnings arise amid an increasing number of incidents involving “rogue” AI agents that have breached sandboxed environments and infiltrated the systems of private companies, alongside concerns that more sophisticated AI tools could be exploited by hackers to access protected networks.
Growth comes with risks
Increased oversight and possible regulatory obstacles could complicate the infrastructure expansion supporting the AI boom. However, Anthropic’s growth has so far defied these worries, driven by soaring demand for its offerings, particularly its popular Claude models.
As of October 31, Anthropic’s run-rate revenue was approximately $81 billion US, which is an increase of $10 billion US since September 30, as per the IPO filing.
The company’s net loss lessened in the third quarter as revenue surged 47% from the previous quarter, reaching $16.96 billion US, although the growth rate has slowed compared to the previous quarter’s 144% surge.
Anthropic attributed its rapid growth to the swift adoption of Claude, including specialized offerings targeting sectors like healthcare, financial services, and legal services.
A smartphone is seen displaying the logos of US artificial intelligence company Anthropic and its Claude AI assistant. Anthropic’s blockbuster initial public offering is set to give the company a valuation of up to $1.4 trillion US. (Imen Ben Youssef/Hans Lucas/AFP/Getty Images)
However, Anthropic’s ambitions come with substantial infrastructure expenses. The company has nearly $900 billion US in computing commitments with partners such as Amazon and Alphabet’s Google, underscoring how its rapid expansion hinges on securing extensive future computing capacity, some of which depends on financing outside the company’s control.
The IPO initiative follows Elon Musk’s SpaceX securing $75 billion US in the largest IPO in history in June, at a valuation nearly $1.75 trillion US, excluding extra shares acquired by underwriters through the greenshoe option.
Along with investments from cornerstone investors like Broadcom, Micron Technology, Microsoft, and TSMC, Anthropic’s share offering aims to raise a total of up to $90.4 billion US.
Certain board members of Anthropic and associated entities have expressed interest in acquiring up to $303 million US worth of shares from the offering.
In a fierce competition for market share against rival OpenAI, Anthropic was last valued at $965 billion US during a private funding round. Its notable investors include Altimeter Capital, Dragoneer, Greenoaks, and Sequoia Capital.
Founded in 2021 by a team of researchers who departed OpenAI due to disagreements over governance and AI safety, Anthropic is led by CEO Dario Amodei.
The shares are anticipated to be traded on the Nasdaq under the ticker symbol ANT once the IPO is finalized, potentially making Anthropic one of the few large firms to go public this fall amid rising borrowing costs, geopolitical instability, and increasing bond yields affecting the IPO landscape.
Goldman Sachs and Morgan Stanley are spearheading the group of investment banks managing the underwriting of the offering.