OpenAI's Codex coding agent - the feature inside ChatGPT that can write, run, and debug code autonomously across a codebase - is now accessible from the ChatGPT mobile app. The update, announced May 14, lets you monitor active tasks, redirect them if they drift off course, and approve next steps, all from your phone.
The workflow this enables is straightforward: start a long-running coding job on your laptop, step away from your desk, and manage it from your phone rather than staying glued to a monitor. OpenAI says the feature works across remote environments too, so if you're running Codex against a cloud-hosted dev environment, the mobile connection follows.
Codex, like most AI coding agents, pauses at decision points to ask for human input before continuing. Previously that meant being at a computer to review the situation and say "yes, proceed." Moving that approval step to mobile is the most genuinely useful part of this change - it shortens the feedback loop without requiring you to be desk-bound.
One practical caveat: approving code changes on a phone screen is a different cognitive experience than reviewing a diff on a monitor. The mobile interface will be most useful for monitoring progress and steering tasks in the right direction, not for catching subtle issues in generated code. For developers running Codex on longer jobs - refactoring a module, writing a test suite, debugging a tricky issue - this reduces the waiting-around problem more than it changes what the agent can do.