OpenAI Codex (AI agent)
Software engineering agent developed by OpenAI
From Wikipedia, the free encyclopedia
Codex is an AI coding agent agent developed by OpenAI and integrated with ChatGPT.[1] It was introduced as a research preview in May 2025, and press coverage of the launch described it as OpenAI's entry into an increasingly competitive market for AI coding tools.[1][2][3] The Verge said OpenAI intended Codex to work more like a coding agent than a conventional assistant, handling tasks independently before returning results for review.[3]
History
OpenAI said Codex was powered at launch by codex-1, a version of its o3 reasoning model optimized for software engineering.[1][3] OpenAI described the initial release as able to write features, answer questions about a codebase, fix bugs, and prepare suggested code changes for review.[1] The research preview first rolled out to some paid ChatGPT tiers, and ChatGPT Plus users gained access in June 2025.[1][4] In February 2026, OpenAI released a desktop Codex app that was intended to help users manage multiple coding agents over longer periods and use code to gather or analyse information.[5]
Features
According to OpenAI, each Codex task runs in a separate cloud environment preloaded with the user's repository, and the agent can read and edit files while running tests and other code-checking tools.[1] OpenAI said most tasks took between 1 and 30 minutes and that Codex returned command logs and test results so users could inspect what it had done.[1] Codex was initially available inside ChatGPT's web app and deliberately lacked general internet access for security reasons.[3] Optional internet access was later enabled during task execution.[1] By February 2026, OpenAI had described a single "App Server" architecture for Codex that powered the CLI, the VS Code extension, the web app, the macOS desktop app, and third-party IDE integrations including JetBrains and Xcode. This design was intended to keep long-running sessions and approval requests consistent across different client interfaces.[6]
In February 2026, Figma announced an integration with Codex that let users move between Figma and Codex through Figma's Model Context Protocol server, connecting interface design and code implementation.[7]
In March 2026, Axios reported that OpenAI had introduced Codex Security, an application-security agent derived from its earlier Aardvark project that was designed to identify, validate, and propose fixes for software vulnerabilities.[8]
Reception
Reuters described coding as one of artificial intelligence's most commercially successful uses and said the market for coding agents had become a key battleground among AI companies.[5][4] In February 2026, Reuters said OpenAI's desktop Codex app formed part of the company's effort to gain ground against Anthropic in the market for AI coding tools, and reported that Codex had been used by more than one million developers in the previous month as OpenAI competed more directly with tools such as Claude Code and Cursor.[5][9] Later Reuters coverage said Codex, along with Claude Code and Cursor, had become a popular way for engineers to quickly create new services.[10] In March 2026, WIRED described Codex as part of OpenAI's effort to catch up with Anthropic in AI coding tools, and reported that the company hoped eventually to use Codex not only for programming but also for task-completing features across its products.[11] In March 2026, Fortune reported that OpenAI said Codex had grown to more than 1.6 million weekly active users after the release of GPT-5.3-Codex, and that companies including Cisco, Nvidia, Ramp, Rakuten, and Harvey had deployed it across developer teams.[12] The same report said OpenAI was presenting Codex as a broader enterprise agent platform that could eventually be used for tasks beyond software development.[12]