Anthropic's record IPO faces some big questions

Anthropic, the rapidly growing AI startup, is on the brink of a record-breaking IPO, valuing the company at $1 trillion. However, the company is facing three major threats that could hinder its success: rival companies catching up with its technology, a collapse in prices, and skyrocketing bills. The prospectus issued by Anthropic outlines these risks and warns that the differentiation among advanced AI models may diminish.

Unlike companies like Google or Meta, Anthropic has no alternative revenue streams to rely on during a downturn. Additionally, open-source AI models are gaining ground, threatening Anthropic's dominance in the market.

So far, investor enthusiasm has waned; a valuation of $2 trillion was floated in mid-August, but growth slowed from 145% in Q2 to 45% in Q3. Open models like Kimi K3.1 Pro, priced half as much as Anthropic's flagship, and others from Anthropic, OpenAI, and Google, are struggling to maintain their positions. Anthropic acknowledges that its safety measures may slow down its technological advancements compared to competitors.

Despite this, new versions of Anthropic's Claude model have boosted sales, with over 80% of its revenue coming from models less than six months old.

The cost of processing AI-generated text has plummeted by 97%, from $17 at the end of 2024 to just 50 cents today. Anthropic itself reduced the price of Claude Opus by 20% in September, and OpenAI cut the price of one of its flagship models by a factor of five in July. Vercel, a developer platform, reports that over half of its text processing is now handled by open models, up from 13% in April. Despite this, Anthropic still captures nearly two-thirds of the spending on the platform.

Anthropic's costs, however, are largely fixed. To maintain its competitive edge, each new Claude model must be larger and more powerful than its predecessor, requiring more data and computing power. The company has committed to spending at least $417 billion on renting chips and servers since late 2025, which could nearly double if cancellable or planned contracts are included.

These commitments are non-negotiable, regardless of the company's financial situation. The International Monetary Fund warns of a "triple whammy" threat if AI proves less profitable than anticipated: falling stock prices, increased debt repayment difficulties, and unfulfilled promised revenue.