OpenAI's $600 million stock sale falters as rival Anthropic surges

A stunning turn of events in the private markets has cast a shadow over OpenAI's dominance, as a reported $600 million block of its shares failed to find a single institutional buyer in early April. The news, coming just as rival Anthropic’s revenue and valuation surge, suggests a significant shift in investor sentiment and intensifies the competitive pressures on the AI pioneer ahead of a potential IPO.

"We实在找不到任何一个愿意接手这些股票的机构投资者 (We simply cannot find a single institutional investor willing to take these shares)," Ken Smythe, founder of secondary market platform Next Round Capital, said in a media interview, noting his firm has hundreds of institutional contacts. He added that buyers told him they had $2 billion in cash ready to purchase shares of rival Anthropic.

The divergence is stark. While OpenAI's secondary shares are reportedly trading at a 10% discount to its last funding round valuation of $852 billion, Anthropic's shares are commanding a premium of over 50% in the same markets. This comes as Anthropic announced its annual recurring revenue (ARR) has soared to $30 billion, eclipsing OpenAI's previously stated $25 billion figure.

This investor pivot complicates OpenAI's path to a public offering and signals that the AI race is entering a new phase. The market's focus appears to be shifting from sheer scale and fundraising prowess to more tangible metrics like enterprise market share and cost efficiency, areas where Anthropic is now perceived to have a decisive edge.

Anthropic Captures 65% of New Enterprise AI Spending

While OpenAI still boasts a massive consumer user base with ChatGPT, the most lucrative enterprise market is rapidly tilting in Anthropic's favor. Ramp data from March 2026 showed that 65% of businesses purchasing new AI services chose Anthropic, compared to just 32% for OpenAI. This momentum is built on a perceived product advantage, with customers reporting that Anthropic's Claude models require less human intervention for complex tasks like drafting legal documents.

Anthropic's dominance is particularly pronounced in key B2B segments. It now commands between 42% and 54% of the code generation market, more than double OpenAI's 21% share. In the enterprise agent market, Anthropic leads with a 40% share to OpenAI's 27%. This enterprise focus accounts for over 80% of Anthropic's revenue, which has grown from an ARR of just $1 billion in January 2025 to $30 billion by April 2026.

This explosive growth has been achieved with far greater capital efficiency. Estimates reported by The Wall Street Journal suggest OpenAI's annual training costs could reach $125 billion by 2030, over four times Anthropic's projected $30 billion. This efficiency advantage suggests Anthropic could reach positive cash flow by 2027, a milestone that remains distant for OpenAI, which has been forced to cut costly projects like the video-generation model Sora.

Global AI Race Resets as New Players Emerge

The cooling sentiment around OpenAI is part of a broader reshuffling in the global AI competition. While OpenAI and Anthropic are now positioned as direct rivals, other major players are carving out distinct strategies. Google is leveraging its vast ecosystem to deeply integrate its Gemini model, focusing on widespread user penetration rather than chasing the "strongest model" title. Meanwhile, Elon Musk's xAI has fallen behind after being acquired by SpaceX and losing most of its founding team.

Perhaps the most significant long-term shift is the rapid ascent of Chinese AI models. On OpenRouter, a major AI model aggregation platform, Chinese models accounted for 12.96 trillion tokens of usage in the last week of March 2026, more than four times the volume of U.S. models. This is a dramatic increase from 2024, when their share was less than 2%. Six of the top ten most-used models on the platform are now from China.

The era of OpenAI's undisputed leadership is over. The failure to sell a major stock block is not merely a secondary market anomaly but a reflection of a new reality. The AI war will not be won by the largest funding rounds or the most ambitious roadmaps alone, but by those who can deliver the most value to customers with the greatest efficiency.

This article is for informational purposes only and does not constitute investment advice.