OpenAI Codex officially lands on mobile! The ChatGPT app can directly write code and review programs

OpenAI Announces Official Launch of Codex AI Code Agent in ChatGPT Mobile App, Enabling Developers to Initiate Tasks, Review Code Outputs, and Guide Execution on iOS and Android, Bringing Professional Development to Mobile Devices and Creating a Truly Mobile Development Experience.
(Background recap: OpenAI’s new engineer agent Codex! AI can write features, fix bugs, run tests... limited to 3 user previews)
(Additional background: Codex mobile operation now live: directly supervise AI remotely in ChatGPT, start new tasks, view outputs)

OpenAI announced via official X account on Taipei time May 21 that its AI code agent tool Codex is officially available in the ChatGPT mobile app, supporting both iOS and Android platforms. This update means developers no longer need to sit at a desktop to write and debug code—just open their phone to start Codex sessions, review outputs, steer execution, and approve next steps anytime.

According to OpenAI’s official statement, the mobile version of Codex is now available in all regions supporting ChatGPT, with Windows users’ mobile connectivity features also soon to be released. This feature upgrades ChatGPT from a chatbot to a true mobile engineering assistant, allowing developers to stay on top of projects during commutes, meetings, or business trips.

Notably, this isn’t Codex’s first interaction with mobile devices. In fact, OpenAI began laying the groundwork for “development anytime, anywhere” for Codex earlier this year—from Chrome extensions enabling Codex to operate on logged-in computers, to backend Mac control and embedded browsers, and now to ChatGPT app integration. This mobile development front is rapidly taking shape.

You've been asking for this one…

Now in preview: Codex in the ChatGPT mobile app.

Start new work, review outputs, steer execution, and approve next steps, all from the ChatGPT mobile app. Codex will keep running on your laptop, Mac mini, or devbox. pic.twitter.com/9i2Jckjt9z

— OpenAI (@OpenAI) May 14, 2026

The Era of Mobile Development Agents Has Arrived: Codex and the Competition

OpenAI Codex landing on mobile directly competes with Anthropic Claude’s app version. While Claude also supports code completion and analysis on mobile, it’s more of a general AI assistant, whereas Codex was built from the ground up as a software development agent—capable of running tests, fixing bugs, and even automatically submitting pull requests.

Meanwhile, Grok, an AI assistant launched by xAI, offers real-time internet info and X platform integration, but still lags behind Codex in pure coding capabilities. Starting from IDEs, Cursor takes a different approach—it’s a desktop-first development environment integration, making it harder to replicate ChatGPT App’s native mobile experience in the short term.

From a developer workflow perspective, the biggest significance of Codex on mobile is breaking the traditional boundary of “development = sitting at a computer coding.” Previously, mobile tools were mainly for reading files, lightweight editing, or CI/CD monitoring; full-fledged code execution and guidance tools for development were rare. Codex’s approach is to bring the complete execution loop of an AI agent—starting tasks, executing, reviewing outputs, guiding next steps, and approving completion—entirely into the phone, allowing developers to maintain control over code quality and progress even when away from their desktops.

Domestic Developer Community Reactions and Opportunities

The domestic software developer community has responded enthusiastically to this update. Several tech groups have shared real-world testing experiences: using Codex on mobile to review GitHub issues and directly generate fixes, then reviewing and approving via ChatGPT app.

“One can only reply to Slack on the subway before; now I can let Codex run test suites, and at the office just check the results—turning commuting time into development time,” said a backend engineer from Taipei participating in testing. Some security developers pointed out that Codex’s output review on mobile is well-designed—AI needs manual approval before executing any operation, balancing convenience and security effectively.

For the vibrant startup ecosystem domestically, the mobile version of Codex could bring more tangible benefits: resource-limited startups often require developers to wear multiple hats. The mobile Codex allows founders to outsource routine development tasks to AI without increasing staffing costs, focusing more on product strategy and business development.

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned