OpenAI launches Codex as powerful new ChatGPT feature while Grok flounders

There’s a weird divide in the AI world these days. While Grok keeps making headlines for all the wrong reasons, OpenAI is pushing ahead with features that actually seem useful for developers. The newest one is called Codex, and it’s built right into ChatGPT for anyone with a Pro, Team, or Enterprise account. Sorry, Plus users.

Codex doesn’t just suggest code, it operates almost like a teammate that can jump into your repository, set up its own isolated workspace, and get to work on whatever you ask. Whether you need a bug fixed, a question answered, or a feature built, Codex can handle it. When it’s done, you get a real code commit and a detailed log showing every step the AI took. How cool is that?

You can also guide Codex by creating an AGENTS.md file in your repo. This is kind of like a cheat sheet for the AI, telling it how things work and what the ground rules are for your project. The more direction you give, the better Codex seems to perform.

What makes Codex feel different from a lot of AI tools is how much transparency it provides. If something breaks or a test fails, it won’t just move on quietly. Everything is right there in the logs. That sort of honesty is refreshing, especially when you compare it to Grok, which lately seems to be stumbling from one controversy to the next.

Codex isn’t limited to the ChatGPT web interface either. There’s a stripped-down command-line version for people who’d rather stay in the terminal, but the real action is inside ChatGPT itself. For now, you’ll need a paid plan, but OpenAI says it’s coming to more users soon.

Codex won’t replace human review, but it brings something real to the table when most AI news feels like empty promises.