Anthropic is weighing whether to release a new AI model ahead of a planned IPO, three people familiar with the company's plans told Reuters, as OpenAI's GPT-6 Astra eats into its enterprise business. The deliberation comes less than two weeks after CEO Dario Amodei published a 3,800-word essay calling on the industry to slow the pace of capability improvements.
The tension is plain. Anthropic built its brand on being the careful one, the company that put safety ahead of shipping speed. Now its own CEO's words are being read back to it while OpenAI's GPT-6 Astra, released September 3, accounts for about 13% of enterprise AI spending tracked by Ramp, against roughly 8% for Anthropic's Claude Fable.
The timing matters. Anthropic's annualized revenue run rate topped $65 billion at the end of July, up from about $9 billion at the end of 2025, and the company has projected 2028 revenue of roughly $190 billion to $200 billion. OpenAI's annualized revenue run rate passed $40 billion in July. With that kind of money on the line, waiting for safety review to finish is a business decision as much as a technical one.
For context on how fast this market moves, see our AI desk coverage of recent model releases. The gap between a safety review and a shipping date used to be measured in weeks. It is now measured in days, because every week of delay hands OpenAI more enterprise logos and more developer habit.
What Astra changed
OpenAI released GPT-6 Astra on September 3, touting gains in computer use, software engineering, cybersecurity and professional work. The model has drawn a strong response from enterprise users and developers, and OpenRouter, a widely used platform that routes developer traffic across AI models, reported that its users spent more on OpenAI models than on Anthropic models last week. That is the first time OpenAI has led on that measure in more than two and a half years.
GPT-6 Astra's 13% share of enterprise AI spending versus Claude Fable's 8%, per Ramp data, is the kind of number that moves investor conversations. Several potential Anthropic IPO investors told Reuters they are re-evaluating Anthropic's position as the leading provider of enterprise AI tools. The company has been viewed for months as the leader in that market.
The safety problem Anthropic can't escape
Amodei's September 12 essay painted a bleak picture of swarms of AI agents taking over the internet and outpacing human control, and he wrote that the industry must slow the pace at which it improves model capabilities. The essay drew support from OpenAI CEO Sam Altman and SpaceX CEO Elon Musk. It was, by any reading, a serious document about a serious risk.
One of the people familiar with Anthropic's deliberations said the company is evaluating the safety of its next model as part of its release decision. That is the crux. Anthropic can either ship now and undercut its own essay, or hold back and watch OpenAI keep pulling ahead in the market it was supposed to own.
Some discussions inside the company involve how to balance investment in releasing new models with efforts to strengthen profitability, given that rising interest rates have made investors more focused on the timing of expected profits. The debate highlights the pressure on AI firms to produce sustainable cash flows sooner, thanks both to rates and to competition from open-source providers, especially in China.
The open-source threat is bigger than the rivalry
Some existing investors and those who expect to invest in both Anthropic's and OpenAI's IPOs said they do not believe Astra poses a significant threat to Anthropic, given the size of its lead in enterprise AI tools and the time it takes to unseat incumbent vendors at large companies. That view is reasonable as far as it goes, but it assumes the fight stays between two companies.
A potentially bigger challenge for Anthropic and other leading providers is the rise of open-source and open-weight models, which can lower token costs and let companies build more of their own AI infrastructure rather than rely on Anthropic or OpenAI. That shift could put pressure on the economics of the whole industry.
Meta Platforms has been among Anthropic's largest customers, but is looking to reduce its use of Anthropic's models as it develops more AI capabilities internally. Meta did not immediately respond to a request for comment. Losing a customer of that size would sting more than any benchmark loss.
Where the IPO stands
OpenAI's CEO confirmed on Saturday that the company would not go public in 2026, citing concerns around AI safety. Anthropic, by contrast, could push its IPO to after the November US midterm elections, with the elections not expected to have a major impact on the offering. Marketing was previously expected to begin in mid-October at the earliest, and the timeline has already slipped.
The potential launch would test how Anthropic can defend its enterprise market position against OpenAI while maintaining the safety-first identity that has distinguished it from competitors. That is the real question, and it is not clear this will hold. A company can be careful, or it can be first. Doing both in a market this fast is a different kind of problem.
According to Reuters, Anthropic declined to comment for the story. Reporting by Echo Wang in New York, with additional reporting by Krystal Hu and Milana Vinn.
The next few weeks will show whether Anthropic ships before the IPO window closes, or whether it lets OpenAI own the enterprise conversation through the fall. Either way, the safety essay is now part of the competitive landscape whether Amodei likes it or not.
Anthropic has not confirmed any launch date, and the company declined to comment. That silence is itself a signal in a market where OpenAI has spent the last month touting Astra's enterprise wins. Investors now have to price in the possibility that Anthropic holds its fire until after the IPO, which would leave the competitive field to OpenAI through the fall and possibly into the holiday quarter.
OpenAI's model announcement and its enterprise push through Google Cloud show how aggressively the company is expanding distribution while Anthropic deliberates.