Anthropic Plans to Disclose $30+ Trillion TAM, Eyeing the Largest IPO in History
Nashnova编辑部
Anthropic plans to tell IPO investors its addressable market tops $30 trillion, surpassing SpaceX's $28.5 trillion record, with a target valuation of $2 trillion and up to $100 billion in proceeds — potentially the largest tech listing ever.
Where does the $30 trillion number come from?
Anthropic's method: count every task an AI model could theoretically perform and sum the economic value. The result is a TAM — total addressable market — of over $30 trillion.
TAM is a hypothetical ceiling — the annual revenue if a product captured 100% market share. In plain terms = it is not a forecast but a thought experiment about how big the sky could be.
For context, the 191 tech companies in the S&P 1500 generated combined revenue of roughly $2.4 trillion last year. Anthropic's claimed TAM is more than 12 times that figure.
NYU valuation authority Aswath Damodaran previously called SpaceX's $28.5 trillion TAM "at the outer limit of credibility." This means → Anthropic's number has crossed that line, and whether the market buys it is the central question.
Is inflating TAM a new trick?
Not at all. Uber cited a $6 trillion market opportunity in its 2019 IPO. WeWork's shelved filing claimed $3 trillion.
SpaceX attributed $26.5 trillion of its TAM to AI, calling it "the largest executable market in human history" — a claim that drew skepticism on Wall Street.
This reflects a pattern: the bigger the TAM, the steeper the credibility discount investors apply. Anthropic must convince the market that $30 trillion is more than a slide-deck number.
Do the revenue numbers back the story?
Q2 revenue hit $11.5 billion, roughly 14× year-on-year. Internal projections put 2028 revenue at $190–200 billion.
The company targets up to $100 billion in IPO proceeds (above SpaceX's $86 billion) and a valuation of roughly $2 trillion (above SpaceX's $1.77 trillion).
A prospectus filing is expected within weeks; the listing could come as early as September or early October. Lead underwriters are Morgan Stanley, Goldman Sachs, and JPMorgan; legal counsel is Wilson Sonsini, the firm behind Google's 2004 IPO.
How deep are the losses, and where does the money go?
Net loss in 2025: roughly $42 billion, about 5× the prior year. In plain terms = revenue grew 14×, but losses grew 5× — the growth is bought with cash.
The spending goes to data centers, chip procurement, and energy — the shared bottleneck for every large-model company.
Adjusted operating profit turned positive in Q2, though durability is unproven. Anthropic is also arranging a $10-billion-plus revolving credit facility to bolster pre-IPO liquidity.
What is different about its chip strategy?
Anthropic runs three chip lines in parallel: Amazon AWS Trainium, Google TPU, and NVIDIA GPU — reducing dependence on any single supplier.
The company has confirmed a custom silicon team and hired Amir Salek — known as the "father of TPU" — from Google to lead it.
This means → Anthropic does not want its compute lifeline in any one vendor's hands; the long-term goal is in-house chip design capability.
Who controls the company after listing?
Anthropic is a public-benefit corporation with a long-term benefit trust whose members include former Fed Chair Ben Bernanke. The trust currently has the power to elect a majority of the seven-seat board.
The company is considering issuing supervoting shares to CEO Dario Amodei and other co-founders. Amodei currently holds roughly 2% of the company; supervoting rights would give him outsized control relative to that stake.
This reflects a tension: a public-benefit charter, a long-term trust, and founder supervoting rights stacked together create three governance layers that public-market investors will need to untangle. SpaceX's stock fell below its IPO price after listing and has only recovered to around $135, near the offering level. This means → no matter how grand the narrative, the secondary market votes with real money.
Content is for reference only, not financial advice.