Macro investor Raoul Pal has publicly criticized Anthropic’s Claude AI model for severe performance issues, warning that the company faces a risk of customer attrition unless it rapidly expands its computing infrastructure.

Pal took to X on Monday to detail his experience, stating that the AI assistant had become difficult to use. He reported wasting hours on tasks that required approximately 30 minutes to complete, describing the service as "utterly unusable."

Performance and Usage Concerns

Pal highlighted specific technical failures in the model’s operation. He noted that the reasoning capabilities were flawed and that the system frequently failed to execute assigned tasks correctly.

The investor also pointed to restrictive usage limits as a barrier. Despite characterizing himself as "hardly a heavy user," Pal stated he exhausts his weekly credits by Monday. He specifically cited issues with the Opus 5 and Fable 5 models, describing the former negatively and complaining about the cost associated with the latter.

"They need more inference and fast or they’re gonna lose a lot of customers," Pal said, urging Anthropic to resolve the bottlenecks quickly.

Regulatory and Competitive Context

This criticism emerges amid broader disputes involving Anthropic’s leadership and market position. CEO Dario Amodei recently pushed back against investor Gavin Baker regarding views on AI regulation. Baker had claimed Amodei suggested Anthropic could eventually become the world’s only private company, a statement Anthropic employee Sholto Douglas called "completely false."

Additionally, Anthropic has faced security allegations in the competitive AI landscape. The company accused Chinese firms DeepSeek, Moonshot AI, and MiniMax of conducting industrial-scale distillation campaigns on Claude involving more than 24,000 accounts and 16 million interactions. Tesla CEO Elon Musk countered these claims by accusing Anthropic of having previously stolen training data at scale.

Anthropic, the maker of the Claude artificial intelligence model, reported that its annualized revenue run rate reached $65 billion by the end of July. The figure represents a sevenfold increase from the same period a year earlier and signals accelerating commercial adoption of its enterprise products ahead of a potential public listing.

The company shared the updated metrics with investors over the weekend, according to reports citing sources familiar with the matter. This latest run rate is also significantly higher than the $47 billion figure Anthropic reported in May, indicating rapid month-on-month growth in its top line.

Revenue Acceleration and Enterprise Demand

In addition to the annualized run rate, Anthropic disclosed preliminary revenue for the second quarter of $11.5 billion. This quarterly figure marks a roughly 14-fold increase compared to the same period in the prior year.

The surge in revenue is driven largely by growing demand from businesses utilizing Claude for coding, research, automation, and other enterprise applications. This performance places Anthropic ahead of rival OpenAI, whose annualized revenue was reported at approximately $40 billion.

What the Numbers Show

The divergence between the Q2 preliminary revenue and the annualized run rate highlights the velocity of Anthropic's recent growth. While the Q2 revenue of $11.5 billion suggests a strong start to the half, the jump from a $47 billion run rate in May to $65 billion in July indicates that monthly revenue growth is outpacing the linear projection of earlier quarters. This acceleration underscores the intensifying competition in the generative AI sector, where Anthropic is now leading in reported revenue scale against key competitors.

IPO Preparations and Valuation

The financial results come as Anthropic prepares for a possible debut on public markets. The company confidentially submitted a draft S-1 registration statement to the Securities and Exchange Commission in June, providing it with the option to go public depending on market conditions.

Anthropic was valued at $965 billion in its latest major funding round, setting high expectations for its initial public offering. However, the company faces ongoing regulatory and government-related challenges, including disputes over the use and export of its AI models. Recent reports indicate that Anthropic projects revenue between $190 billion and $200 billion by 2028.