• Anthropic reported a $567 million operating loss for the quarter ended September 30, 2026, signaling that rapid sales growth has not yet translated into consistent profitability.
  • The AI company filed its IPO registration statement on October 26 and plans to sell up to $3.5 billion of shares to Japanese investors, according to Reuters (TRI).
  • The loss follows a $974 million operating profit in the second quarter, underscoring the volatility of AI economics as computing costs mount.

A Stumble Before the Bell

Anthropic’s $567 million operating loss for the September quarter, if confirmed, would mark a sharp reversal from the $974 million operating profit it posted in the prior three months. The figure, contained in a headline summary, has not been independently verified in regulatory filings reviewed by this publication. Nevertheless, it highlights the challenge facing the San Francisco-based AI lab as it transitions from private high-flyer to public-market contender.

The company’s revenue momentum remains formidable. According to reported investor disclosures, September-quarter revenue reached approximately $16.8 billion, up about 45% from the previous quarter. For the first half of 2026, revenue totaled roughly $16.3 billion, a fourteenfold increase from $1.2 billion a year earlier. Yet operating losses in that same period narrowed to $693 million from about $2.1 billion, suggesting that profitability was beginning to take hold—until the latest quarter’s setback.

IPO Machine in Motion

Anthropic filed its S-1 registration statement with the SEC on October 26 and intends to list on Nasdaq (NDAQ) under the ticker “ANT.” As part of the offering, the company plans to allocate up to 552.9 billion yen, or $3.5 billion, of shares to Japanese investors, Reuters reported on Monday. Proceeds will fund working capital, operating expenses, and capital expenditure, according to the report.

The IPO comes at a delicate moment. Investors are increasingly scrutinizing whether AI revenue can justify the enormous infrastructure spending that underpins the sector. Anthropic has committed more than $600 billion to computing deals, a multiyear obligation that will keep pressure on margins. Its prospectus flags dependencies on a limited group of cloud providers—Amazon (AMZN), Google (GOOGL), and Microsoft (MSFT)—as a vulnerability.

“What institutional investors like us are really focused on is regulatory stability,” said Andrea Valeri, Blackstone (BX)’s country chairman for Italy, speaking at a separate event. His comment, while not specific to Anthropic, reflects the broader demand for predictability as AI companies seek public capital.

The Profitability Puzzle

Anthropic’s financials reveal a company still finding its footing. The first-half net loss of approximately $80.7 billion was largely driven by a $72.4 billion non-operating accounting charge related to revaluing convertible instruments—a paper loss, not a cash drain. Adjusted operating income for the first half was $1.62 billion, a figure that paints a rosier picture than standard accounting.

But the September-quarter loss suggests that operating leverage remains elusive. Computing and infrastructure expenses are expected to rise in absolute dollars and remain the company’s largest operating cost, management has warned. Revenue growth may also slow, adding further uncertainty.

“It’s a great country to invest here because there are a lot of very good companies and the market here is not as competitive as other markets,” said Giampiero Mazza, head of Italy at CVC Capital Partners (CVC.AS), in an unrelated interview. His comment underscores the competitive intensity that Anthropic faces globally—not just from other AI labs but from a market that is growing more selective.

Governance and Government

Anthropic operates as a public benefit corporation and is classified as a “controlled company” because its Long-Term Benefit Trust can elect a majority of the board. The IPO disclosure named Netflix (NFLX) co-founder Reed Hastings as lead independent director. On October 21, the company announced that physicist Christopher Stubbs had joined the trust, and a week earlier it unveiled a new Public Benefit Fund, committing a portion of its shares to that entity.

Government business is also expanding. On October 27, Anthropic announced an expanded General Services Administration agreement offering a 40% discount on Claude Enterprise, Claude for Government, and its API. The move could broaden public-sector adoption, though the revenue impact remains unclear.

Internationally, the company announced a Singapore office and a government memorandum of understanding on October 14 to advance responsible AI adoption. The planned Japanese share allocation would connect Japanese capital more directly to the U.S. AI expansion.

Market Reaction and Outlook

The IPO is being watched as a bellwether for the entire AI sector. IPOX’s Lukas Muehlbauer told Reuters that Anthropic’s listing will establish a benchmark for leading AI laboratories, while Renaissance Capital’s Matt Kennedy described a more challenging valuation environment than during the summer. Investors are reportedly discussing a valuation around $1.5 trillion, down from roughly $2 trillion a week earlier—a sign of shifting sentiment.

The next consequential disclosures will be the full September-quarter financial statements and the reconciliation between operating and adjusted results. If the $567 million loss is confirmed, investors will need to determine whether the reversal from second-quarter profit reflects temporary capacity additions or a more persistent cost problem.

Anthropic did not respond to a request for comment on the operating loss figure.

Correction: An earlier version of this article misstated the period of the $693 million operating loss. It covers the first six months of 2026, not the September quarter.