Anthropic is considering the release of a new artificial intelligence model as OpenAI’s GPT-6 Astra gains traction among businesses and developers. According to three people familiar with the matter cited by Reuters, the company is weighing a potential launch as it prepares for an expected initial public offering (IPO).
The discussions come at a notable time for Anthropic. CEO Dario Amodei recently called for a slower pace of AI development because of concerns about the risks of increasingly capable models. At the same time, the Anthropic new AI model discussion puts the company under pressure as OpenAI’s latest model gains ground in the enterprise AI market.
GPT-6 Astra Gains Enterprise Traction
OpenAI releases GPT-6 Astra on September 3 and highlights improvements in computer use, software engineering, cybersecurity, and professional work. The model receives strong interest from businesses and developers, according to the Reuters report.
Recent spending data also shows increased adoption. Corporate expense platform Ramp tracks GPT-6 Astra at about 13% of enterprise AI spending, while Anthropic’s Claude Fable accounts for about 8%. These figures give investors another measure of how quickly businesses adopt newer AI models, while the Anthropic new AI model discussion adds to the competition.
OpenAI also gains ground on OpenRouter, a platform that routes developer requests across different AI models. OpenRouter reports that users spend more on OpenAI models than on Anthropic models during the latest week covered by the report. This marks the first time OpenAI leads Anthropic on that measure in more than two and a half years.
The spending data does not establish that businesses are broadly replacing Anthropic with OpenAI. Enterprise AI contracts often involve long-term deployments, testing, security reviews, and integration work. Some investors therefore continue to view Anthropic as having a substantial position with large businesses.
Anthropic Weighs a New Model
The reported discussions center on whether Anthropic should release an Anthropic new AI model in response to OpenAI’s recent momentum. However, the company has not announced a launch date, and the sources do not indicate that a final decision has been made.
Anthropic is also evaluating the safety of its next model, according to one person familiar with the matter. This consideration is especially relevant because the company continues to emphasize AI safety in its model development and public communications. Anthropic declines to comment on the reported plans.
The timing also creates a contrast with Amodei’s recent public position. In a September 12 essay, he calls for the AI industry to slow the pace at which it improves model capabilities. He raises concerns about increasingly capable AI agents and the difficulty of keeping them under human control. A new model release would therefore require Anthropic to balance competitive development with the safety processes that it publicly emphasizes.
Profitability Adds Another Consideration
The decision is also connected to the rising cost of developing frontier AI models. Training and operating advanced models require large investments in computing infrastructure, research, data, and safety testing.
According to the Reuters report, Anthropic is discussing how spending an Anthropic new AI model fits with its efforts to improve profitability. Higher interest rates also increase investor attention on when major AI companies can generate sustainable returns from their large investments.
Anthropic still reports significant business growth. Its annualized revenue run rate exceeds $65 billion by the end of July, compared with about $9 billion at the end of 2025. The company also projects revenue of roughly $190 billion to $200 billion in 2028. OpenAI’s annualized revenue run rate passes $40 billion in July.
These figures show that competition between the two companies involves more than model benchmarks. Enterprise spending, revenue growth, infrastructure costs, customer retention, and the ability to turn AI demand into sustainable business all play a role.
Open Models Create Another Challenge
The competition surrounding the Anthropic new AI model and OpenAI is also taking place alongside the growth of open-source and open-weight AI models. Some investors cited in the Reuters report see these models as a broader challenge because they can reduce token costs and allow companies to build more AI infrastructure internally. Businesses that have the technical resources to run their own models gain another option besides relying entirely on commercial AI providers.
Meta also represents an example of this shift. The company remains one of Anthropic’s major customers but is reportedly looking to reduce its use of Anthropic models as it develops more AI capabilities internally. This development highlights how large technology companies can simultaneously use commercial models and invest in their own AI systems.
What Comes Next for Anthropic
Anthropic’s reported consideration of an Anthropic new AI model shows how quickly the enterprise AI market changes as new model generations arrive. GPT-6 Astra’s recent launch gives OpenAI additional momentum among some businesses and developers, while Anthropic continues to maintain strong revenue growth and enterprise relationships.
The company now weighs several factors at once: competitive pressure from OpenAI, safety evaluations, development costs, profitability, and its expected IPO. There is no confirmed decision on whether Anthropic releases the new model or when such a model could arrive.
For enterprise customers, the developments mean the AI model market continues to expand and change quickly. Companies have more choices across commercial, open-weight, and internally developed systems. As Anthropic, OpenAI, Google, and other AI developers release new generations, businesses continue to evaluate models based on their capabilities, costs, security requirements, and suitability for specific workloads.
