Anthropic has filed its initial public offering prospectus with US regulators, disclosing $16.3 billion in first-half revenue and plans to list on Nasdaq under the ticker ANT.
The filing, submitted to the Securities and Exchange Commission on Monday, October 26, offers investors a detailed assessment of the artificial intelligence developer’s financial performance, infrastructure obligations and governance arrangements ahead of its proposed stock market debut.
Revenue for the six months ended June 30, 2026, reached $16.286 billion, compared with $1.163 billion during the corresponding period of 2025. The increase reflects accelerating commercial adoption of Claude, particularly among businesses integrating artificial intelligence into software development and corporate operations.
However, the prospectus also disclosed a net loss of $80.748 billion, substantially exceeding the $12.969 billion loss recorded a year earlier. The extraordinary figure was driven largely by non-cash accounting adjustments associated with convertible financing instruments rather than equivalent operating cash expenditure.
Approximately $72.4 billion of the loss reflected changes in the assessed value of convertible notes, highlighting how financing structures can produce substantial accounting charges as private technology companies increase in valuation.
Anthropic’s operating loss narrowed to approximately $693 million from $2.1 billion in the comparable period, suggesting improving underlying performance despite the enormous expenditure required to develop and operate advanced artificial intelligence systems.
Computing and infrastructure expenses totalled approximately $8.945 billion during the first half, against $2.33 billion a year earlier. Nevertheless, those expenses declined significantly as a proportion of revenue, from approximately 201% to 55%.
The company also recorded approximately $974 million in operating profit during the second quarter, demonstrating that revenue growth has begun to outpace certain operating expenses, although profitability across the full six-month period remained elusive.
The scale of Anthropic’s infrastructure commitments represents another significant disclosure. Agreements covering computing capacity, cloud services and associated infrastructure involve hundreds of billions of dollars in potential expenditure extending across several years.
Identified arrangements with seven computing providers encompass approximately $637 billion in minimum commitments and potential additional payments, although the contractual terms distinguish binding obligations from expenditure dependent on capacity delivery and utilisation.
Further prospective arrangements involving specialised processing equipment could increase the overall investment substantially. Such figures do not represent immediate liabilities or expenditure already incurred, making their contractual classification important for assessing future financial exposure.
Anthropic identifies computing capacity as its principal operating cost, reflecting the enormous processing requirements involved in training increasingly capable models and serving customers through Claude applications and programming interfaces.
The prospectus indicates that usage-based services generated approximately 80% of first-half revenue, exposing the business to fluctuations in customer demand while infrastructure agreements can impose longer-term payment obligations.
Customer concentration presents another consideration. Two customers collectively represented approximately one-fifth of first-half revenue, creating potential exposure should either reduce spending or transfer workloads to competing providers.
The company’s commercial expansion has nevertheless produced exceptional customer retention figures. Anthropic disclosed net revenue retention of approximately 900% as of September 30, reflecting substantial increases in spending among existing customers.
Such growth demonstrates the rapid expansion of artificial intelligence usage within established accounts, although the company cautioned that its performance might not translate into sustained profitability.
Anthropic has not specified the number of shares it intends to offer or their proposed price range. Consequently, the filing does not establish a definitive market valuation or the amount of capital the flotation will raise.
Goldman Sachs is leading the proposed offering, with Morgan Stanley assigned a stabilisation role. The transaction remains subject to regulatory procedures and prevailing market conditions.
The proposed shareholding arrangements would preserve substantial influence for Anthropic’s founders. A special voting structure is designed to provide the seven co-founders with 50.1% of voting power on matters submitted to shareholders generally.
Separately, the company’s Long-Term Benefit Trust would retain authority over appointments to a majority of authorised board positions, reflecting Anthropic’s governance framework for balancing commercial objectives with artificial intelligence safety considerations.
The prospectus also identifies technological reliability, cybersecurity, regulatory intervention and possible misuse of advanced models among material business risks.
Anthropic, founded in 2021, is led by chief executive Dario Amodei, while co-founder Daniela Amodei serves as president. Both are identified in the registration documents as principal company representatives for the proposed offering.
Related Posts
Add A Comment