Anthropic's revenue rises 45% to $16.8 billion in Q3 ahead of IPO

Anthropic reported a 45 per cent rise in revenue in the third quarter to $16.8 billion from $11.5 billion in the previous quarter, as the artificial intelligence (AI) company sought to build investor confidence ahead of a potential initial public offering (IPO) as early as November.

At the meeting in San Francisco on Wednesday, Chief Executive Officer Dario Amodei and Chief Financial Officer Krishna Rao outlined the company's financial performance, growth plans and approach to AI safety. The event was intended to familiarise prospective investors with the business ahead of its anticipated listing.

The Financial Times reported that the Claude chatbot maker told investors it had been profitable on an adjusted basis for the second consecutive quarter. The company also highlighted its cash generation and improving computing efficiency as it prepares for a public listing amid growing scrutiny of the cost of developing and running AI models.

The pace of growth has slowed from the more than doubling of revenue between the first and second quarters. The report said the latest figures were also below some investors' earlier expectations, with several backers having anticipated annualised revenue of more than $100 billion by the end of the year.

Research firm SemiAnalysis estimated that the company's annualised revenue had reached about $74 billion based on the latest quarterly numbers, the Financial Times reported. According to a report by Reuters, a person who attended the investor presentation said the company was expecting an annualised revenue run rate of $110 billion by the end of the year.

According to the Financial Times, Anthropic told investors it was cash-flow positive in the third quarter and generated $1.70 in revenue for every dollar spent on computing power. Its adjusted operating margin stood at 11 per cent.

Competition and costs cloud growth outlook

Anthropic faces growing competition from OpenAI and cheaper Chinese AI models, while some major customers have increasingly turned to their own AI tools.

The Financial Times reported that Meta and Microsoft, both significant Anthropic customers, have expanded their use of in-house AI tools. The company has also reduced the contribution from model distillation, a practice in which developers train AI systems using the outputs of more advanced models.

Anthropic told investors that distillation accounted for 5-7 per cent of its revenue and that it had cut back the activity materially in the third quarter, the Financial Times reported.

The company and OpenAI have also reduced prices as they compete for customers, adding to the pressure to improve efficiency even as demand for AI services grows.

Anthropic has committed at least $518 billion over a decade to cloud services, computing and infrastructure, Reuters reported. The rising interest rates and a more challenging macroeconomic environment could constrain plans for heavy investment in computing capacity.

Anthropic is planning to get listed after the November US midterm elections. The company is also seeking to differentiate itself through its emphasis on AI safety. The Financial Times reported that a substantial portion of an early version of its draft IPO filing was devoted to risk factors, including the possibility of unpredictable or malicious behaviour by AI models.

The company has warned investors that its technology could pose “existential risks to humanity”, underscoring the challenge of balancing commercial expansion with concerns over the risks associated with increasingly capable AI systems.