Anthropic is studying the possibility of releasing a new artificial intelligence model to counteract the growing influence of OpenAI following the launch of GPT-6 Astra. This move is being considered against the backdrop of the company's expected public offering and after Anthropic's CEO called for a slowdown in the industry's pace of development.
Sources reported that the decision on a potential launch aimed at mitigating competitive pressure from a direct rival was made after Anthropic's head, Dario Amodei, addressed the global AI community with a call to reduce the speed of releasing new features for safety issues.
In his 3800-word essay from September 12, Amodei presented a grim picture of a scenario where swarms of AI agents could take over the internet and surpass human control. His theses found support from OpenAI CEO Sam Altman and SpaceX CEO Elon Musk.
OpenAI's GPT-6 Astra, released this month, has gained recognition among corporate clients. This has raised concerns among some investors who have begun to reassess Anthropic's position as a leading provider of enterprise AI tools.
One informed source stated that Anthropic is assessing the safety of its next model during discussions about its release. Among the topics under discussion are how to balance investments in releasing new models with efforts to strengthen the company's profitability, as rising interest rates cause investors to focus more on timelines for expected profits.
Capturing Corporate Demand for Astra
Discussions about spending highlight the pressure on firms at the center of the AI boom to achieve sustainable cash flow quickly. This pressure is driven by both interest rates and fierce competition from open-source AI providers, especially from China.
Anthropic declined to comment. A potential launch would allow Anthropic to test how it can defend its position in the corporate market against OpenAI while maintaining its safety-focused identity, which distinguishes it from competitors.
OpenAI released GPT-6 Astra on September 3, advertising improvements in computer usage, software development, cybersecurity, and professional work. The model received a strong response from corporate sector users and developers, prompting potential Anthropic IPO investors to closely examine whether OpenAI might start taking market share from Anthropic, which had been considered the leader in enterprise AI for months.
According to the latest data, Astra accounted for about 13% of enterprise AI spending tracked by the corporate spending platform Ramp, compared to approximately 8% for Anthropic's Claude Fable. Furthermore, Astra surpassed OpenRouter, a widely used platform routing developer traffic between AI models. OpenRouter reported that users spent more on OpenAI models than on Anthropic models last week, marking the first time OpenAI has led in this metric in over two and a half years.
Some existing investors and those expecting to invest in both Anthropic and OpenAI IPOs believe that Astra does not pose a significant threat to Anthropic, given the size of its leadership in the enterprise AI tool market and the time required to displace established providers from large companies.
Anthropic's annual revenue reached over $65 billion by the end of July, up from approximately $9 billion at the end of 2025. The company also forecasts revenue of between $190 and $200 billion for 2028. OpenAI's annual revenue exceeded $40 billion in July.
Investors also expect that leadership among Anthropic, OpenAI, Google, and other major AI developers will frequently change as companies release new generations of models, making any advantage potentially short-lived.
A more serious problem for Anthropic and other leading AI providers is the rise of open-source and open-weight models. Such models can lower token costs and allow companies to build much of their own AI infrastructure instead of relying on providers like Anthropic and OpenAI, investors note. This shift could put pressure on the AI industry's economics, as companies gain more options for developing and deploying models outside of leading commercial providers, expanding the competitive threat beyond the race between Anthropic and OpenAI.
Meta Platforms was one of Anthropic's largest clients, but according to informed sources, it plans to reduce its use of Anthropic models as it develops more internal AI capabilities itself. Meta did not respond immediately to the request for comment.
Further IPO Delays
Meanwhile, OpenAI has eased some pressure on the race to public markets. Altman confirmed on Saturday that the company will not go public in 2026, stating that AI safety concerns make it an unsuitable time for a listing.
According to two informed sources, Anthropic may postpone its IPO until after the US Congressional midterm elections in November, although the election is not expected to have a significant impact on the listing. The IPO has already been delayed compared to initial plans; Reuters previously reported that marketing was not expected to begin before mid-October.