OpenAI CEO Sam Altman outlined a three-part business strategy at the company’s developer conference in San Francisco: offer competitive models, lower the barriers to building applications, and develop OpenAI into a platform connecting enterprise customers with partners. As ChatGPT’s user base expands, the plan shows how OpenAI is seeking to turn model capabilities into broader developer and enterprise revenue.
Altman announced updates to ChatGPT and related products, describing the strategy as “models, building and distribution.” OpenAI and rival Anthropic are both under pressure to demonstrate that their high valuations can be supported by durable commercial growth. Google, Meta, SpaceXAI and Chinese model developers are also competing for technical capability and customers.
GPT-6.1 Sol targets multiple price points
Altman said models remain the starting point for OpenAI’s strategy. The company began as a research laboratory but now operates in a market where competitors continue to improve performance while using lower prices to attract customers.
At the conference, he introduced GPT-6.1 Sol as a more affordable option alongside the company’s latest Astra model. OpenAI also launched its Ultrafast service, which offers faster model access at a significantly higher price. Altman said the company wants to provide products that balance intelligence and functionality across different price points in the market.
The approach would allow OpenAI to cover a wider range of use cases through different models and access options. Competition in the model market, however, will continue to focus on both performance and cost, as Anthropic and other providers keep developing their products.
Codex moves to the cloud as OpenAI adds agent APIs
The second part of the strategy is to make it easier for developers to build products using OpenAI technology. Altman said the company wants to make tools it uses internally available to developers.
OpenAI announced that its Codex programming tool will be extended to the cloud, allowing developers to use it across different devices. The company also released an application programming interface for customers building agents and introduced another API designed for situations requiring rapid decisions. The latter announcement drew a strong response from the audience. TypeSafe AI’s Jev model offers a similar capability.
Altman described the initiative as a one-stop building platform for developers. For OpenAI, tools and APIs can expand the use of its models while making it easier for companies and developers to embed the technology in existing products and workflows.
ChatGPT’s user base supports an enterprise platform
The third part of the strategy is distribution and connectivity. OpenAI said ChatGPT has more than 1.2 billion weekly users, giving the company a large base from which to expand its enterprise and partner businesses. At the same time, more companies want to integrate OpenAI technology or use its products to reach customers.
Altman said enterprise customers could use part of their existing OpenAI commitments to purchase products from partners. Users could also select “Sign in with ChatGPT” on other AI products and use part of the token allocation associated with their accounts. These arrangements would move OpenAI beyond its role as a model supplier and toward becoming an intermediary for enterprise services and product distribution.
During a question-and-answer session following the keynote, Altman used cloud-service providers as a reference point when discussing how OpenAI could support a broader ecosystem. He said the company wants to help customers connect with products and direct users, revenue and customers toward partners, creating a more integrated network.
The plan places model competition, developer tools and customer distribution within a single commercial framework. Its ability to expand enterprise revenue and partner businesses will depend on customer adoption and on whether OpenAI can continue meeting different requirements for model pricing, performance and service coverage.