'Price War' Anthropic: Will Its Growth Story Collapse After Going Public?

[Financial News]  


Clouds are gathering over Anthropic, the developer of the Claude AI model, which is expected to stage the largest IPO ever. Fierce competition among AI models is raising questions about whether the company can sustain its current growth story. Warnings are growing that its stock could plunge after surging in the wake of a blockbuster IPO.
FT reported on the 19th (local time) that investors are questioning whether Anthropic can maintain its extraordinary growth even after its IPO.

For now, the mood is upbeat.
Annualized revenue rose to $65 billion in July, and optimists expect it to nearly double by year-end, surpassing $120 billion. The company is projected to post the steepest growth ever achieved by any business.
However, significant uncertainty has fueled concerns that this growth story could collapse.
Investors value Anthropic at $965 billion, or approximately 1,340 trillion won. Its valuation is widely expected to rise further after the listing, reaching between $1.5 trillion and $4 trillion. That would put its market capitalization on par with Amazon's.

Skepticism toward the rosy outlook is also growing. The biggest reason is the strong performance of Anthropic's largest rival, OpenAI.
OpenAI is reclaiming market share after unveiling its new AI model, GPT-5.6, in July. According to OpenRouter, Anthropic's weekly customer spending fell behind OpenAI's for the first time in two and a half years after the new model was released.
The competition does not end there. Both OpenAI and Anthropic are facing threats from 'open' models. These are AI models whose parameters are publicly available and can be customized.
These open models are led by Chinese companies such as DeepSeek and Moonshot AI, as well as Meta Platforms. They are cheaper than Anthropic's and OpenAI's offerings. Their performance is also improving rapidly, and they can now rival the 'frontier' models developed by OpenAI and Anthropic. As a result, their market share is growing.
Eric Glyman, co-founder of corporate payment systems company Ramp, said, "This is the first time there has been real price competition among AI companies."
AI customers, including Ramp, switch among models depending on the task. There is no need to use expensive Anthropic or OpenAI models for simple tasks. Ramp cut its related spending by 40% by using different AI models.

Glyman emphasized, "You don't need to rent a Ferrari to go grocery shopping."
This is gradually becoming a reality, as industry figures including Microsoft CEO Satya Nadella have long predicted. They have said that AI models will eventually become commoditized, triggering price-cutting competition, and that users will choose the model offering the best value for money.
Mike Pollus, a former partner at legendary Silicon Valley venture capital firm Andreessen Horowitz, said, "If AI works well and a business model takes hold, Anthropic will be at the center of it," while noting that competition is unavoidable. He stressed that OpenAI and numerous open-model AI companies will compete fiercely alongside Anthropic, adding, "The AI market is not a cozy monopoly."

Expectations that AI, like the software industry in its early days, could secure high margins in a competition-free environment are also weakening. Pessimism is intensifying even before the company goes public, without a honeymoon period. Fierce competition has already sparked a price war, making exclusive margins difficult to achieve.
There are also concerns that AI companies such as Anthropic could be influenced by major customers, cloud partners and the Donald Trump administration.
Concerns about human extinction have also begun to hinder efforts to improve profitability in the AI industry. Anthropic researcher Jacob Coxon resigned from the company on the 8th, urging developers to slow down after saying, "AI developers genuinely believe that AI could kill us all by 2030."
Anthropic CEO Dario Amodei also called for slowing down development on the 11th. His rivals, OpenAI CEO Sam Altman and SpaceX CEO Elon Musk, unusually voiced support for the proposal.
Slowing down can reduce costs, but companies must also be prepared for the technology gap with competitors such as China to narrow. That is not what the market wants.
Pollus noted, "While AI CEOs are saying, 'We need to slow down and be cautious,' the market is essentially responding, 'That's true, but the incentive to pursue profits is overwhelmingly strong.'" He added, "We may one day regret not taking their—the warning voices'—words at face value."

[email protected] Song Kyung-jae Reporter