Anthropic has been profitable for two consecutive quarters ahead of its IPO

Anthropic is currently at a critical juncture before going public, presenting an impressive financial report to the market as proof of its viable business model.

According to the Financial Times on September 13, sources revealed that the Claude large model developer has informed some shareholders that its adjusted operating profit will be positive for the second consecutive quarter this quarter, marking a significant milestone for this five-year-old AI unicorn as it charges towards an IPO. At the same time, Anthropic has selected Nasdaq as its listing venue, and the IPO could be valued at $2 trillion or even higher.

On the financial data front, Anthropic’s performance is quite convincing. According to reports, sources said the company’s gross margin—before deducting revenue-sharing with distribution partners like Amazon and model training costs—has already exceeded 80%; second-quarter revenue soared 14 times year-on-year to $11.5 billion, while annualized revenue reached $65 billion by the end of July—a huge leap from $9 billion at the end of last year. Investors expect Anthropic’s annualized revenue will reach $120 billion by year-end, and could triple that by the end of 2027.

However, the road to IPO is not entirely smooth. Just before the IPO, Anthropic CEO Dario Amodei publicly called for the AI industry to slow the pace of capability improvements, a stance that resonates with OpenAI and Elon Musk, prompting potential investors to reassess the possible impact of stricter AI safety regulations on the company’s growth trajectory.

Consecutive Profitability, a Significant Financial Milestone

According to reports, multiple sources revealed that Anthropic has disclosed to a small group of shareholders that adjusted operating profit will be positive this quarter, marking its second consecutive quarter achieving this metric. This figure excludes non-cash costs such as equity incentives.

This achievement is highly significant for a cutting-edge AI company sprinting toward an IPO. One of the long-standing doubts about AI large model companies is whether their significant computing investments and ongoing cash burn can translate into sustainable commercial returns. Anthropic’s disclosure of consecutive profitability directly addresses this skepticism.

SemiAnalysis analyst Joey Brookhart stated:

"If Anthropic can continue to maintain this kind of profit margin and growth rate, competitors will find it extremely hard to catch up, as Anthropic has already accumulated massive computing power resources."

Valuation May Reach $2 Trillion, Nasdaq Bell Ringing Approaches

According to reports, sources revealed that Anthropic has selected Nasdaq as the listing venue, and the potential valuation of this IPO is $2 trillion or higher. The company is currently valued at about $965 billion.

In terms of the listing process, Anthropic originally planned to publicly file its prospectus last week, but sources said the company has instead chosen to share documents and accept inquiries privately with a small group of investors before public disclosure. This means the official listing schedule remains somewhat flexible.

By comparison, competitor OpenAI secretly filed for IPO in June, but its CEO Sam Altman confirmed to Fortune magazine on Saturday that OpenAI will remain private this year, stating that an IPO in 2026 would be "unwise" as debates over AI safety have not yet subsided.

Industry Calls for Slowdown, Safety Concerns Cast Shadow Over IPO Outlook

At this crucial point for Anthropic’s IPO push, a rare internal industry dialogue is quietly reshaping external public opinion.

According to the report, sources said in recent weeks, staff at several leading AI labs have been in private, close communication to discuss how to safely manage AI development. Such conversations are very rare in this highly competitive industry, and are driven by factors including recent safety incident breaches, growing unease among researchers regarding the capabilities of new-generation AI models, and the widespread belief that the Trump administration is unlikely to proactively slow down technological progress. One frontier lab employee said bluntly: "We can only rely on ourselves."

These private discussions laid the groundwork for Amodei’s and Altman’s public statements this week. In an article published Saturday, Amodei made it clear that the AI industry "must slow the pace of AI model capability improvements" because the powerful capabilities of new models have sparked widespread alarm. Altman and Musk quickly echoed this stance publicly.

Analysts believe this position brings a dual narrative: While slowing the development pace may save Anthropic billions of dollars in model training costs, it could also give competitors an opening to close the gap. How Anthropic balances regulatory pressure for safety and commercial competition will be one of the core narrative challenges it faces after going public.