OpenAI brings Codex to mobile ChatGPT, putting dev control on your phone
OpenAI has brought Codex to the mobile ChatGPT app. From an iPhone or Android smartphone, you can now start and continue tasks and view diffs, terminal output and test results while the agent works on a connected Mac. From the phone you can also approve commands, answer the agent's questions and switch between threads. The feature launches in preview for all plans, including Free and Go. Windows support is promised for later.
AI-processed from The Verge; edited by Hamidun News
OpenAI is extending Codex beyond the desktop: the tool for programming and managing tasks is now available in the mobile ChatGPT app. Users can start or continue work from their phone while Codex itself executes commands on a connected computer.
How access works
The new mode launches in preview on iOS and Android. In effect, ChatGPT on a smartphone becomes a remote interface for Codex, which keeps running on a MacBook, a dedicated Mac mini or a remote development environment. Screenshots, terminal output, test results, diffs and requests to confirm actions arrive on the phone in real time.
The feature is available on all ChatGPT plans, including Free and Go, but only in supported regions. OpenAI stresses separately that files, credentials and local permissions are not transferred to the smartphone. They stay on the machine where Codex is running, and the devices communicate through a secure relay layer.
To get started, you need to update the ChatGPT app and the Codex app on macOS, then pair the devices with a QR code. The host must stay powered on, connected to the network and keep Codex open. For now, the phone can be connected only to Codex on macOS; the company promises to add Windows support later.
What you can do from your phone
It is not just a "run task" button. OpenAI describes mobile Codex as a full working surface: you can switch between threads, answer the agent's clarifying questions, change the model, approve commands and start new tasks without returning to your computer. This is especially useful for long workflows where the agent reaches a fork and waits for a human decision. For a mobile interface, this is no longer an auxiliary feature but a way to keep your pace of work.
- Start a bug investigation and watch Codex reproduce the error and run tests
- Approve the next command or pick one of the refactoring options on the road
- Review the diff and test results before getting back to your desk
- Ask Codex to put together a short incident brief before a call with a client
"Now you can stay in the context of active work from anywhere," is how
OpenAI describes the idea behind mobile access to Codex.
According to the company's official data, Codex is already used by more than 4 million people a week. The mobile version is meant to close an obvious gap: if an agent works for tens of minutes or longer, the user needs more than a desktop interface; they need a way to step into the process quickly without interrupting what they are doing to open a laptop. This matters especially for teams whose development already relies on remote devbox environments and long automated pipelines. Otherwise even a strong agent too easily ends up waiting for a human reply.
Why OpenAI is doing this
The launch looks like a logical follow-up to the recent Codex update for macOS, in which the tool learned to work with applications on the computer. Now OpenAI is adding another layer: remote supervision of those tasks from a phone. In practical terms, the company is building not a standalone "code generator" but a broader work environment around ChatGPT and Codex. It also strengthens ChatGPT's position as a central interface for working with agents, not just for ordinary chat.
Against this backdrop, the move also looks like a response to intensifying competition in AI development. After the surge of interest in Claude Code, OpenAI noticeably sped up Codex development and, judging by its latest releases, is focusing on products that can be built into the daily workflow of developers and enterprise teams.
The company also says Remote SSH is already available on all plans, so Codex can run not only on a local Mac but also inside managed corporate environments with the required dependencies, security policies and compute resources.
What this means
AI development tools look less and less like a chat with suggestions and more and more like a permanent worker to whom a human periodically gives direction. If mobile Codex proves stable in real-world use, the phone will become not a spare screen but a proper control point for long engineering tasks.
- Nov 1, 2026Local models lose to Opus and Codex: 0.21 on a MacBook versus 1.00 for Codex
- Nov 1, 2026Kimi scores 96.2% on a 1C database analytics test and emerges as a fallback if Claude access is blocked
- Nov 1, 2026OpenAI Rewrites GPT Behavior Rules: Five Values and a Chain of Command
- Nov 1, 2026Google admits AI Mode cuts search revenue, shares fall 15%
Key Questions Answered
How access works?
The new mode launches in preview on iOS and Android. In effect, ChatGPT on a smartphone becomes a remote interface for Codex, which keeps running on a MacBook, a dedicated Mac mini or a remote development environment. Screenshots, terminal output, test results, diffs and requests to confirm actions arrive on the phone in real time. The feature is available on all ChatGPT plans, including Free and Go, but only in supported regions. OpenAI stresses separately that files, credentials and local permissions are not transferred to the smartphone. They stay on the machine where Co
What you can do from your phone?
It is not just a "run task" button. OpenAI describes mobile Codex as a full working surface: you can switch between threads, answer the agent's clarifying questions, change the model, approve commands and start new tasks without returning to your computer. This is especially useful for long workflows where the agent reaches a fork and waits for a human decision. For a mobile interface, this is no longer an auxiliary feature but a way to keep your pace of work. Start a bug investigation and watch Codex reproduce the error and run tests Approve the next command or pick one of
Why OpenAI is doing this?
The launch looks like a logical follow-up to the recent Codex update for macOS, in which the tool learned to work with applications on the computer. Now OpenAI is adding another layer: remote supervision of those tasks from a phone. In practical terms, the company is building not a standalone "code generator" but a broader work environment around ChatGPT and Codex. It also strengthens ChatGPT's position as a central interface for working with agents, not just for ordinary chat. Against this backdrop, the move also looks like a response to intensifying competition in AI deve
Build & Ship Production AI Agents Without Boilerplate Code
Start your free 72-hour trial at Hamidun Academy (academy.hamidun.com) — interactive lessons on Claude Code & Cursor, live Tech Radar, and production agent templates.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.