Key Points
- Anthropic’s rapid growth has slowed, cooling expectations for a valuation of up to $1.5 trillion.
- Falling AI prices and cheaper open models threaten the company’s ability to sustain profit margins.
- At least $417 billion in computing commitments raises questions about future cash flows and financial risks.
ISLAMABAD: Anthropic is preparing for a potentially record-breaking stock market debut, but its rapid rise in artificial intelligence faces a profitability test.
Other challenges and questions include competition, rising infrastructure costs and eroding investor confidence.
The San Francisco-based company, which developed the Claude family of AI models, has grown into one of the world’s most valuable private technology businesses in five years.
Its prospectus, filed ahead of a planned Nasdaq listing, outlines the risks to its ambition to turn growing demand for AI into sustainable returns.
The proposed offering could value Anthropic at around $1.5 trillion, though some investors expect a $2 trillion valuation. The listing is expected in November and could become one of the largest stock market debuts in history.
AI competition threatens growth
Anthropic’s first challenge is intensifying competition from established tech companies and developers of open AI models that users can download and modify rather than access exclusively through paid services.
The company warns in its prospectus that differences between advanced AI models could narrow, making it harder to justify premium prices.
Models from Anthropic, OpenAI and Google have repeatedly competed for the performance lead, leaving no provider with a lasting advantage across all categories.
Open models are also closing the performance gap. Research institute Epoch AI estimates that their capabilities trail those of leading closed models by about four months on average.
One competitor, China’s Moonshot, offers its Kimi K3.1 Pro model at roughly half the price of Anthropic’s flagship Claude Opus 5.5, intensifying pressure on customers to compare performance against cost.
Anthropic also acknowledges that its emphasis on safety could slow its technical progress relative to competitors that impose fewer constraints.
The company must therefore balance safeguards, product development and the need to maintain its competitive position.
Prices fall as costs climb
Pricing presents another challenge. Morningstar estimates that the cost of processing the same amount of text at comparable performance has fallen 97 per cent, from $17 at the end of 2024 to 50 cents now.
Anthropic cut the price of Claude Opus by 20 per cent in September, while OpenAI reduced the price of one of its flagship models fivefold in July.
Lower prices could expand AI adoption by making the technology affordable to more businesses.
But they also threaten margins unless companies can offset declining revenue per unit through greater usage, improved efficiency or additional services.
Anthropic argues that expanding demand can absorb substantial price reductions.
Morningstar equity analyst Malik Ahmed Khan said overall spending on AI could grow even if prices fell sharply, provided usage maintained its current pace.
READ ALSO: Anthropic Revenue Reaches $16.8b Ahead of Planned Listing
The question is whether that growth will suffice to cover the industry’s rising costs.
These commitments reflect the enormous computing resources required to train capable models and serve customers.
Safety and financial risks
The prospectus also highlights risks beyond pricing and infrastructure. Anthropic disclosed that it had paused pre-training its next-generation flagship model for more than a month to strengthen protections against the theft of model weights and intellectual property.
The company also reported cybersecurity testing incidents in which four AI models escaped their testing environments and accessed external computer systems.
Such incidents could increase regulatory scrutiny, raise development costs and undermine confidence in AI companies if safeguards fail.
Anthropic also depends heavily on its AI products for revenue, unlike technology groups with established advertising or cloud-computing businesses that can provide additional income streams.
Google and Amazon are simultaneously its investors, computing suppliers and distribution partners. This complex financial link could become a source of vulnerability if AI returns disappoint.
The International Monetary Fund has warned that interconnected investments in AI infrastructure could amplify financial stress through falling asset prices, mounting repayment difficulties and revenues that fail to materialise.
A test for public investors
Anthropic’s listing will test whether public-market investors are prepared to finance the next phase of AI development at valuations built on expectations of sustained growth.
Its prospectus points to substantial opportunities, but the company must demonstrate that newer models can generate enough additional demand to offset falling prices and rising computing costs.
