The filing reveals a dramatic surge in financial activity over the past two years. According to the document, Anthropic’s revenue jumped from $386 million in 2024 to nearly $4.6 billion in 2025. The growth accelerated sharply in 2026, with the company reporting revenue of nearly $16.3 billion in the first six months alone. This represents a significant leap from the $1.1 billion generated in the same period of the previous year. However, this revenue growth has been accompanied by substantial financial losses. Anthropic reported a net loss of $80.7 billion in the first half of 2026, a steep increase from the $13 billion loss incurred in the same period a year earlier.
The primary driver of these losses is the company’s infrastructure expenditure. The prospectus identifies compute and infrastructure expenses as the largest operating cost for the foreseeable future. These costs have escalated rapidly, rising from $433 million in 2023 to over $2.5 billion in 2024, and further surging to $7.3 billion in 2025. In the first half of 2026, compute and infrastructure expenses reached more than $8.9 billion, compared to $2.3 billion in the first half of 2025. This trajectory reflects the immense capital requirements associated with training and deploying large-scale AI models, a burden shared by other frontier labs but highlighted here by the sheer scale of the figures.
Despite the losses, Anthropic is approaching the public markets with a high valuation. Earlier this year, private investors valued the company at $965 billion after a $65 billion funding round in late May. Reports suggest the IPO could see this valuation rise to between $1.5 trillion and $2 trillion. At the lower estimate, Anthropic would rank among the world’s most valuable public companies, comparable to Tesla. The company plans to list its stock on the Nasdaq Global Select Market under the ticker symbol “ANT.” The exact amount of capital to be raised and the final share price range were not disclosed in the filing, though reports indicate the offering could raise as much as $100 billion, which would set a new record for an IPO.
The timing of the filing coincides with a period of market volatility and heightened scrutiny on the AI sector. While the announcement provided a brief boost to U.S. stock index futures, investor sentiment remains cautious. Financial analysts have noted that the high valuation relative to current losses poses a risk. Thomas Hayes, Chairman at Great Hill Capital, described a $1.5 trillion valuation for a company losing $80 billion to $160 billion annually as “rich,” suggesting that the IPO is primarily for investors who believe in the long-term return of capital. Others, such as Josh Stevens of Cresalta Investment Management, warned that great business models often see share prices correct to more attractive levels during periods of reassessment, advising caution rather than immediate participation.
Anthropic’s financial disclosures also highlight its heavy reliance on a limited number of customers and distribution channels. The company’s early strategic partnerships with Amazon and Google have been crucial, with these tech giants providing both billions of dollars in investment and the cloud infrastructure necessary to train and deploy Claude models. In its prospectus, Anthropic asserted that it believes the “abundance of AI” will transform every sector of the global economy, citing an impact potentially greater than industrialization, electricity, semiconductors, or the internet.
The IPO filing comes amidst broader debates within the AI community regarding the pace of development. CEO Dario Amodei recently argued in an essay for slowing the development of the most advanced models, citing potential existential risks. This caution contrasts with the aggressive capital expansion implied by the IPO. The company’s ability to sustain its current burn rate while scaling its infrastructure will be a key metric for public investors to monitor as the offering progresses. The final outcome of this IPO is expected to serve as a significant benchmark for the sentiment surrounding capital expenditures in the artificial intelligence sector, influencing how other frontier labs, including rival OpenAI, approach their own potential public listings.
