Artificial intelligence firm Anthropic has reported a net loss exceeding $80 billion for the first six months of 2026, according to a report by the Wall Street Journal. The figures, which emerged from a draft prospectus intended for financial regulators, indicate a substantial increase from the $13 billion loss recorded during the same period in 2025. This financial disclosure comes as the company, recognized for developing the Claude chatbot, prepares for a highly anticipated initial public offering (IPO) expected in November.
The primary driver behind the significant loss was $72 billion in accounting charges. These expenses are reportedly linked to fluctuations in the value of convertible notes and various other financial instruments that Anthropic had previously issued to secure necessary capital. Beyond these paper losses, the company revealed a strategy for future expansion that could involve spending as much as $750 billion in the coming years to significantly bolster its computing capacity. These plans reportedly involve potential acquisition deals that remain under active discussion.
The financial data provides a rare window into the intense fiscal demands of the current artificial intelligence race. Investors are increasingly scrutinizing whether the massive capital being diverted into high-performance chips and expansive data centers will generate the expected long-term returns. While revenue for the six-month period ending June 30 climbed to $16.3 billion—up from $1.1 billion during the same window the previous year—the cost of scaling infrastructure has risen in tandem, with infrastructure spending nearly quadrupling to approximately $9 billion.
In the documentation shared with regulators, the company outlined several risk factors pertinent to its business model. Among these is the ongoing uncertainty regarding how to mitigate potential harms caused by its technology, including the risks of sophisticated cyberattacks or the misuse of uncontrolled models. Furthermore, the company reported last week that its artificial intelligence systems could potentially be only months away from gaining the capability to self-replicate on standard home computing devices, adding another layer of complexity for potential investors to consider.
The latest disclosures appear to have influenced market sentiment. Following an investor day held in San Francisco, California, last week, some current and prospective investors are reportedly weighing a lower valuation for the firm. People familiar with the discussions indicated that stakeholders are now looking at a valuation of approximately $1.5 trillion, which represents a notable decline from the $2 trillion figure that had been previously projected for the upcoming share sale.
Anthropic, which has gained market traction through its coding assistant, Claude Code, did not respond to requests for comment regarding the reports. As the company moves toward its public market debut, the discrepancy between its rapid revenue growth and the escalating costs of AI development remains a central point of discussion among industry analysts and financial stakeholders. The reliance on significant infrastructure investment continues to be a defining characteristic of the firm's growth trajectory as it navigates the transition from a private entity to a publicly traded corporation.
The figures reported by the Wall Street Journal arrive on the heels of other financial disclosures, including a Reuters report from September noting that the company posted a net loss of $42 billion in 2025. As the November share sale approaches, market participants are expected to continue scrutinizing these prospectus documents to gauge the sustainability of Anthropic's ambitious spending on computing resources against its evolving revenue streams.
Comments