Anthropic: 9 Risks Bears are Focusing оn Ahead of the IPO

While the upcoming Anthropic IPO is creating an immense amount of excitement within the investment community, not everyone is bullish on the AI powerhouse. Independent research firm New Constructs, for example, recently published a short-seller report on Anthropic in which it argued that the company's target valuation of $2 trillion is vastly inflated and that the IPO is an unprecedented test of investor gullibility¹. 

Here, we are going to highlight nine risk factors that the Anthropic bears are focusing on ahead of the IPO. From the sky-high valuation to the company’s reliance on cloud service providers (CSPs), these risk factors form the foundation of the bear case.  

The Sky-High Valuation   

Let’s start with the valuation. With analysts expecting Anthropic to command a market cap of $1.5 trillion to $2.0 trillion², the AI lab is coming to the market with a price target that leaves little room for error.

At a $2.0 trillion market cap, the company would be sporting a price-to-sales ratio of around 30, taking the ARR figure of $67 billion³ that has been reported recently. Taking last year’s actual revenue figure of $4.6 billion⁴, the price-to-sales ratio would be a staggeringly high 436.     

A Lack of Profitability

Moving past the valuation, another key factor the bears are focusing on is Anthropic’s lack of profitability. In 2025, the AI company generated an operating loss of $8.1 billion⁴.  

Now, operating losses did shrink to $693 million in the first half of 2026. However, the company has warned that spending on compute could hurt its profits in the future.   

Enormous Compute Costs

Speaking of compute, Anthropic’s costs here continue to soar. In the first half of 2026, compute and infrastructure costs amounted to $8.9 billion⁴.

Note that in the company’s Form S-1 prospectus, it said that subsequent to December 31, 2025, it had entered agreements to secure additional compute and infrastructure capacity, with aggregate future minimum payment obligations of approximately $417 billion⁴. It also said that if it is unable to secure adequate compute, it may be unable to provide market-leading AI models.    

Model Commoditization

On the growth side, one of the reasons Anthropic is having so much success today is that it has a market-leading AI model in Claude. The problem is that competitors are quickly narrowing capabilities gaps.

Looking ahead, enterprises could turn to AI models from the likes of OpenAI and Google to reduce their reliance on a single company. They could also turn to open-source or open-weight models to lower their AI costs.     

Customer Concentration

Zooming in on Anthropic customers, its top customers are responsible for a large chunk of its overall revenue today. In its S-1, it revealed that for the nine months ended September 30, 2026, its top five customers represented approximately 25% of total revenue⁴.

It added in the S-1 that many of its top customers are not subject to long-term contracts. So, there is definitely some customer concentration risk here – if one or more of its top customers were to reduce their spending, the company could be significantly impacted.  

Reliance on Cloud Service Providers  

A related risk here is the company’s reliance on cloud service providers (CSPs) such as Amazon and Google for distribution. For the year to June 30, 2026, CSP marketplace revenue represented 48% of total revenue⁴.

The issue here is that the CSPs have their own AI models. So, there is a chance that they could prioritize their own products over Anthropic’s models in the future.    

AI Model Risks

Turning to its products, Anthropic noted in its S-1 that if the use of its AI models results in harm – whether through use by malicious actors or misuse by other third parties – its reputation, financial strength, and operational results could be impacted negatively. It added that it could be exposed to liability and/or litigation arising from harms associated with its AI models.

Furthermore, it stated that its AI models could behave in unintended, misaligned, or autonomous ways. In the past, its models have exhibited unexpected strategic or self-preserving behaviors, including attempts to resist shutdown, conceal or manipulate information, or generate outputs that could be interpreted as coercive, deceptive, or manipulative, including behavior resembling blackmail. 

Political and Regulatory Risks

Political and regulatory risks also can’t be ignored. In its S-1 filing, Anthropic warned that its public positions, stated values, and policy advocacy have attracted – and may continue to attract – criticism from influential political figures, industry participants, and regulators.

It went on to say that scrutiny or backlash could result in reputational harm, increased compliance or operating costs, investigations, enforcement actions, or restrictions on its operations. All of these scenarios could adversely affect its business, financial performance, and results. 

Access to Capital

Finally, access to capital is another major risk that the bears are focusing on. In its S-1, Anthropic said that it may require additional capital to meet its substantial ongoing obligations and support business growth.

However, other tech companies such as Alphabet, Meta, SpaceX, and OpenAI are also looking to raise capital and not everyone believes there is enough capital to fund every organization’s AI spending plans. Broadcom CEO Hock Tan, for example, recently said that financing could become a bigger obstacle to AI expansion than shortages of semiconductors, memory chips, or electricity⁵.  

The Bear Case Has Merit  

In summary, while Anthropic's enterprise market dominance and growth trajectory have captured the public's imagination, the risk factors above highlight the significant challenges lying ahead. A sky-high valuation, massive infrastructure costs, and intense competition mean that strong hype does not automatically guarantee long-term market success.

Investors should remember that even the most promising technology companies can face volatile market dynamics and complex operational headwinds at times. Ultimately, an upcoming public listing offers no sure path to profitable returns, making thorough due diligence essential before jumping in. 

Footnotes:

¹New Constructs, Anthropic Is The Most Ridiculous IPO of 2026, as of October 6, 2026 

²Investing.com, Anthropic files for IPO, with revenue surging and potential $2 trillion valuation, as of October 26,2026 

³ARR Club, Anthropic, as of October 29, 2026 

⁴Sec.gov, 1FORM S-1 REGISTRATION STATEMENT, Anthropic, as of October 26, 2026 

⁵CNBC, Intel, Broadcom CEOs warn capital crunch could become AI’s biggest bottleneck, as of October 12, 2026