Anthropic's third-quarter revenue came to $16.8 billion, up 45 percent from the previous quarter. It also recorded an adjusted profit for a second straight quarter. With an initial public offering (IPO) ahead, it has shown both growth and profitability, drawing attention to listing plans that target a valuation of up to $2 trillion.

SiliconANGLE reported on Oct. 14 that Anthropic disclosed the results in private meetings with investors interested in buying shares after a listing. The market expects Anthropic to pursue an IPO in November.

Anthropic's revenue rose to $11.6 billion in the second quarter from $4.8 billion in the first quarter, and increased to $16.8 billion in the third quarter. It was not confirmed whether the figure was gross revenue or net revenue, and the company did not disclose in detail what 기준 it used to calculate profitability. Investors are focusing not only on revenue growth but also on the costs spent to achieve it.

As a private company, Anthropic has no obligation to disclose detailed financial information. For that reason, even limited disclosures can affect the broader tech stock sector. Recently, after reports that OpenAI's annualised revenue estimate was lowered to $50 billion from about $68 billion, the Nasdaq index and shares of related companies including Oracle, CoreWeave, Intel and Nvidia fell. These companies have large long-term contracts with artificial intelligence (AI) firms, closely tying their business performance together.

The competitive landscape in the AI market is also changing quickly. Unlike OpenAI, once considered a leading player, Anthropic has expanded its foothold by releasing state-of-the-art AI models in succession. Growing corporate demand for Claude Code, which generates code from natural-language instructions, has also supported that growth. Anthropic completed a $65 billion fundraising in May, and was valued at $96.5 billion. Operating high-performance AI models requires large amounts of capital to cover the cost of expensive computing chips.

But controversy over AI safety remains a burden. In September, Anthropic researcher Jacob Croxon (제이컵 크록슨) left the company and said AI developers "seriously believe that AI could kill all of us by the end of the decade." After that, Anthropic Chief Executive Dario Amodei (다리오 아모데이) urged that the industry "should slow the pace of improvements in AI model capabilities" so it can secure time to put safeguards in place. Sam Altman and Elon Musk also agreed, but President Trump supports companies' voluntary self-regulation.

Price competition is also intensifying. As concerns grow over the cost of AI services, more companies are seeking relatively cheaper open-source and open-weight models. In response, Anthropic released the small language model Claude Haiku 5.5 last week and said its cost was about 75 percent lower than the previous model, Haiku 4.5. Competition with OpenAI's small language model GPT-6 Luna also appears to be ramping up.

Anthropic's third-quarter results show AI demand is continuing, but they also highlight the costs needed for growth, safety issues and pressure to cut prices. Investors are expected to watch whether the revenue gains translate into a sustainable profit structure and whether Anthropic can maintain its competitive edge.