Is the AI leader changing hands? Rumor has it that Anthropic might raise funds again at a $900 billion valuation, surpassing OpenAI's valuation.

Cladue developer Anthropic is expected to raise funds at a valuation of $900 billion, surpassing OpenAI to become the leader in AI. Benefiting from soaring revenue, the company is accelerating its IPO plans, with a listing possibly as early as October this year.

Anthropic valuation may reach $900 billion, challenging OpenAI

Claude developer Anthropic’s valuation could surpass that of ChatGPT developer OpenAI? According to TechCrunch, sources reveal that Anthropic has received multiple preemptive investment offers, with investors proposing a valuation between $850 billion and $900 billion, and raising an additional $40 billion to $50 billion.

Anthropic is expected to make a decision at the board meeting in May 2026. If the deal goes through, its valuation will double and surpass the current OpenAI. Previously, OpenAI disclosed in a press release that it had completed a $122 billion funding round, with its valuation rising to $852 billion.

Recently, Business Insider and Bloomberg reported that some venture capitalists have offered Anthropic $800 billion. In February, Anthropic completed financing led by Singapore’s sovereign wealth fund and Cotu Capital, with a valuation of $380 billion.

Meanwhile, on the private secondary market Caplight, Anthropic’s implied valuation has risen to $688 billion, a 75% increase in three months.

Image source: Flickr/Getty Images/Kimberly White Photography Anthropic CEO Dario Amodei participating in an event hosted by TechCrunch in 2023

Anthropic revenue surges, accelerating IPO plans

Anthropic’s annual recurring revenue (ARR) has surpassed $30 billion, a significant increase from $9 billion at the end of 2025, now approaching $40 billion. Currently, over 1,000 enterprise clients spend more than $1 million annually, with data doubling in two months. The growth is mainly driven by Claude Code and Cowork platforms.

Bloomberg previously reported that the strong revenue generated by Claude-related products has prompted Anthropic to prepare for its first public offering, with a potential listing as early as October this year. Meanwhile, OpenAI also revealed monthly revenue reaching $2 billion, with over 40% from enterprise clients, and continues to expand market share through GPT-5.4 models.

OpenAI recently launched GPT-5.5, focusing on scientific research and coding, and has shut down the money-losing AI-generated video app Sora. However, the new GPT Image 2 model has reignited public interest due to its more detailed image generation and greater editability.

• Related article: How to use GPT Images 2.0? Practical tests on beef noodle menus, magazine covers, and multilingual science explanations

Mythos model draws attention, cybersecurity regulation becomes a focus

Anthropic also has a mysterious trump card—the “Mythos model.” Due to its ability to identify software vulnerabilities, it is currently not publicly available to avoid cybersecurity risks, only provided to select vendors and organizations.

However, on April 21, Bloomberg reported that unauthorized organizations had illegally obtained and used the Mythos model. The White House and Pentagon have launched an emergency inter-agency investigation into the leak, focusing on the increasingly fragile AI software supply chain.

Meanwhile, the U.S. White House and Anthropic maintain a nominal blacklist relationship. Sources within the intelligence community reveal that after Mythos was listed on the NSA’s internal security blacklist, the NSA still maintains access rights to the tool.

• Related article: The mysterious AI model Mythos leaks! Illegal organizations have deployed copies—has Anthropic’s security defenses been breached?

As AI competition heats up, the rivalry between Anthropic and OpenAI has extended from mere technological prowess to the deep waters of national security and commercial applications.

In the future, who can truly secure the throne as AI leader may no longer depend solely on valuation size or model computational limits; how to balance advancing cutting-edge technology with building trust among enterprises, regulatory oversight, and safeguarding fragile software supply chains will be the ultimate key to winning this AI technology race.