OpenAI’s annual developer conference on Monday was a spectacle of showcases for ambitious AI products, from: app store for ChatGPT to stunning Video generation API that brought artistic concepts to life. But for closely watching enterprises and tech leaders, the most significant announcement was the silence general availability of the CodeAI software engineer. This release signals a profound change in the way software, and therefore current business, is created.
While other announcements have captured the public’s imagination, the release is ready for production Codesupercharged by A new, specialized model ia enterprise-class toolsetit is the engine of the entire OpenAI vision. It is a tool that builds tools, a proven agent in a world teeming with agentic potential, and the clearest expression of a company’s strategy to win the business.
The general availability of the Code takes it from a “research preview” to a fully supported product, along with the up-to-date one software development kit (SDK)AND Slack integrationand administrative security and monitoring controls. This transition is a declaration that Codex is ready for mission-critical work in the world’s largest companies.
“We believe this is the best time ever to be a developer; getting from idea to product has never been faster,” said OpenAI CEO Sam Altman at the conference opening thought presentation. “It used to be that software development took months or even years. Now you’ve seen that software development using AI can take minutes.”
This acceleration is not theoretical. This is a reality born out of OpenAI’s internal exploit, a massive solution testing effort that serves as the ultimate case study for enterprise customers.
Inside GPT-5-Codex: an artificial intelligence model that codes autonomously for hours and delivers a 70% enhance in productivity.
The heart of updating the Code is GPT-5 Codea version of OpenAI’s latest flagship model that has been “purposefully trained for Codex and agentic coding.” The up-to-date model is designed to function as an autonomous member of the team, going far beyond straightforward code autocomplete.
“Personally, I like to think of it as a human teammate,” explained Tibo Sottiaux, an OpenAI engineer, during a technical session on the Code. “You can pair a program with him on your computer, you can delegate it to him, or, as you’ll see, you can give him a task without an explicit prompt.”
This up-to-date model enables “adaptive thinking”, allowing you to dynamically adjust the time and computational effort spent on a task based on its complexity. For simple requests it is fast and efficient, but for complex refactoring projects it can run for hours.
One engineer noted during a technical session, “I saw the GPT-5-Codex model run productively for over seven hours… during a marathon session.” This ability to handle long-term, complex tasks is a significant step up from the simple, one-off interactions that define most AI coding assistants.
The results within OpenAI were dramatic. The company reported that 92% of its technical staff now use Codex every day, and these engineers complete 70% more pull requests (a measure of code contribution) every week. Since August, usage has increased tenfold.
“When we see the stats as a team, it makes us feel great,” Sottiaux said. “But it’s even better to have lunch with someone who then says, ‘Hey, I exploit the Code all the time. Here’s a frigid thing I do with it. Want to hear about it?'”
How OpenAI uses Codex to build its own AI products and catch hundreds of bugs a day
Perhaps the most compelling argument for the importance of the Code is that it is the foundational layer on which OpenAI’s other flashy announcements are built. While DevDay eventfor the conference itself, the company presented custom-made arcade games and a dynamic AI-powered website, all developed using Code.
In one session, engineers demonstrated how they built “Storyboard,” a custom creative tool for the film industry, in just 48 hours during an internal hackathon. “We decided to test Codex, our coding agent… we submitted tasks to Codex between meetings. We found it really basic to review and connect PRs to production, which Codex even allowed us to do from our phones,” said Allison August, solutions engineering lead at OpenAI.
This reveals a key insight: the rapid innovation showcased at DevDay is a direct result of the productivity flywheel created by Code. AI is a key part of the manufacturing process for all other AI-based products.
A key feature aimed at enterprises is the new, more robust code review feature. OpenAI said it has “intentionally trained GPT-5-Codex to be great at very thorough code review,” enabling it to examine dependencies and check developer intent against the actual implementation to find high-quality bugs. Internally, almost every pull request on OpenAI is now reviewed by Codex, catching hundreds of issues a day before they reach a reviewer.
“It saves time and provides greater confidence for ships,” Sottiaux said. “There’s nothing worse than finding a bug after the feature has been released.”
Why enterprise software teams choose Codex over GitHub Copilot for mission-critical development
Ripening Code is central to OpenAI’s broader strategy to conquer the enterprise market, a move necessary to justify its massive valuation and unprecedented compute spend. During the press conference, CEO Sam Altman confirmed the strategic change.
“The models are already there, and the expectation is that we will focus on the really winning ventures with amazing products, starting from here,” Altman said during a private news conference.
OpenAI CEO and co-founder Greg Brockman immediately added: “You can already see that with Codex, which I think has been incredibly successful and has grown really quickly.”
For technical decision-makers, the message is clear. While consumer-facing agents who book dinner reservations are still finding their footing, Code is a proven corporate agent providing significant ROI today. Companies like Cisco have already implemented the Code within their engineering organizations, reducing code review times by 50% and reducing project turnaround times from weeks to days.
With the new one SDK Codecompanies can now embed this agent power directly into their own custom workflows, such as automating patches in the CI/CD pipeline or even building self-evolving applications. During a live demonstration, the engineer demonstrated a mobile application that updated its own user interface in real time based on natural language messages, all powered by the built-in Codex SDK.
While commissioning application ecosystem in ChatGPT and breathtaking visuals API Sister 2 correctly generated headlines, general availability of the Code it means a more fundamental and immediate transformation. It is the peaceful but powerful engine powering the next era of software development, turning the abstract promise of AI-powered productivity into a physical, implementable reality for today’s businesses.