Anthropic targets $2tn valuation to eclipse SpaceX’s IPO

Anthropic is preparing to ask public-market investors to value the artificial intelligence company at more than $2 trillion, setting up a test of whether the market will place a frontier AI developer above SpaceX’s record-breaking public debut despite dramatically lower revenue and much heavier operating losses.

The Claude maker’s proposed valuation would more than double the $965 billion valuation assigned to Anthropic in its May 2026 funding round and surpass the $1.77 trillion valuation at which Elon Musk’s SpaceX priced its June initial public offering.

SpaceX raised $75 billion in its IPO at $135 a share, making it the largest US IPO on record by proceeds. Its shares subsequently pushed the company’s market value above $2 trillion shortly after listing.

Anthropic’s proposed valuation therefore puts two very different businesses on almost the same financial stage: a diversified space and satellite-internet company with tens of billions of dollars in annual revenue, and a five-year-old AI laboratory whose commercial growth is accelerating but whose cost of computing is rising almost as quickly.

The comparison is significant because Anthropic generated $4.59 billion in revenue in 2025, up 1,088 percent from $386 million a year earlier, according to its IPO prospectus reviewed by Reuters. Yet the company recorded a net loss of about $42 billion and an operating loss of $8.06 billion.

About $34 billion of the net loss came from a non-cash accounting charge linked to the revaluation of convertible financing instruments as Anthropic’s private valuation increased. That means the headline loss does not represent $42 billion of cash leaving the company.

But stripping out the accounting charge does not make the underlying economics comfortable. Anthropic still lost more than $8 billion from operations in 2025 as it spent heavily on computing capacity, infrastructure and the development of increasingly capable AI models.

The company spent $7.33 billion on compute and infrastructure during the year, accounting for more than half of its $12.65 billion in operating expenses.

More importantly for prospective shareholders, Anthropic has disclosed plans involving about $518 billion in future commitments for cloud computing and infrastructure.

That commitment is more than 100 times the company’s 2025 revenue, highlighting the capital intensity of the frontier-AI business and the scale of the bet investors would be making through the proposed listing.

Read also: Airtel Money to reveal London IPO price range in October

SpaceX offers investors a different financial proposition. The company generated $18.67 billion in revenue in 2025, roughly four times Anthropic’s revenue, while reporting a $4.94 billion net loss. Its business spans Starlink satellite connectivity, rocket launches and other space activities, giving the company multiple sources of revenue rather than a single frontier-AI product.

SpaceX’s June IPO was priced at $1.77 trillion and raised $75 billion, setting a record for the largest US initial public offering.

That makes Anthropic’s planned valuation particularly notable. At more than $2 trillion, Anthropic would be asking investors to attach a higher value to a company whose 2025 revenue was about one-quarter of SpaceX’s and whose operating loss was substantially larger.

The difference is not simply a question of current financial performance. SpaceX entered public markets with Starlink already operating at scale and with a large space business alongside it. Anthropic is effectively asking investors to value the future economics of Claude and the wider frontier-AI market.

That future is where Anthropic believes the valuation can be justified.

Revenue growth is providing the foundation for the argument. A jump from $386 million to $4.59 billion in one year demonstrates how rapidly enterprise demand for AI services can expand. Reuters reported that the company is expected to pursue a valuation of about $2 trillion, making the listing one of the most consequential tests yet of public-market appetite for AI.

But revenue growth alone does not resolve the central question facing investors: how much of every additional dollar of AI revenue will ultimately remain as profit after paying for chips, cloud capacity, electricity, data centres and model development?

Anthropic’s financial disclosures show that question is becoming more important, not less.

The company’s $518 billion in future infrastructure commitments illustrates how the economics of AI differ from those of traditional software.

A conventional software company can add millions of customers without necessarily increasing its infrastructure costs at the same rate. Frontier AI models require enormous computing resources both to train and to serve users.

As adoption grows, Anthropic must therefore acquire more computing capacity to support that growth.

This creates a difficult financial equation. Stronger demand can produce rapidly rising revenue, but stronger demand can also require more computing infrastructure.

Anthropic is betting that its revenue curve will eventually rise faster than its infrastructure costs.

The proposed $2 trillion valuation effectively asks investors to believe that this transition from high-growth, high-burn AI laboratory to highly profitable technology platform can happen at extraordinary speed.

That is a much bigger assumption than the headline $42 billion loss suggests.

Anthropic is also entering the public markets with significant customer concentration.

Two customers accounted for nearly a quarter of its 2025 revenue, while some large customers are not tied to long-term contracts.

That matters because a substantial valuation requires investors to assume that today’s rapid growth can continue across a much broader customer base.

If enterprise customers increase their use of Claude, Anthropic can spread its infrastructure investments across a growing revenue base. If customer spending proves more volatile, the company’s enormous infrastructure commitments could become a greater burden.

The proposed valuation consequently rests on several linked assumptions: continued rapid AI adoption, sustained revenue growth, improving economics per unit of computing and the ability to secure enough infrastructure to meet demand.

Anthropic’s own risk disclosures add another unusual dimension to the public offering.

The company devotes about 80 pages of its 261-page core prospectus to risk factors, including warnings that increasingly capable AI systems could create “catastrophic or existential risks to humanity”.

It also warns that advanced models could potentially exhibit self-preserving behaviour, resist shutdown, manipulate information or engage in behaviour resembling blackmail.

For a company whose valuation depends on the continued development and commercial deployment of increasingly powerful models, those warnings create an unusual tension.

Anthropic is telling investors that its technology could become one of the most economically transformative technologies ever developed while simultaneously acknowledging that more capable systems could introduce risks that are difficult to predict or control.

The issue is therefore no longer confined to AI safety researchers. It becomes part of the investment proposition for public shareholders.

Read also: Simple Access to the Dangote Refinery IPO with i-invest

Anthropic’s proposed $2 trillion valuation is ultimately a bet on the future economics of artificial intelligence rather than its 2025 financial statements.

The company’s revenue growth gives investors evidence that demand for frontier AI is expanding rapidly. But its $8.06 billion operating loss and $518 billion of future infrastructure commitments show the cost of capturing that demand is also enormous.