OpenAI rolls out Codex, an AI coding agent powered by codex-1, a version of o3 optimized for software engineering, for ChatGPT Pro, Enterprise, and Team users
OpenAI announced on Friday it's launching a research preview of Codex, the company's most capable AI coding agent yet.
TechCrunchMaxwell Zeff
Context & Ripple Effects
This moves Codex from its earlier role as a natural-language-to-code API, which powered GitHub Copilot, toward a productized agent inside ChatGPT’s paid business tiers. It follows the April introduction of Codex CLI for terminal-based local tasks, extending the product surface from the command line to Pro, Enterprise, and Team users.
The significance is distribution as much as model capability: OpenAI is placing an engineering-focused o3 variant in the same subscription environment used by teams, creating a clearer path from assisted code generation to delegated software-engineering work.
First-order effects
ChatGPT Pro, Enterprise, and Team customers gain research-preview access to Codex, while OpenAI gains a direct channel to test and refine an agentic coding product with paid users.
Software teams can evaluate a specialized coding agent alongside existing ChatGPT workflows rather than adopting a separate coding API or terminal tool.
Second-order effects
Coding-assistant vendors and model providers face greater pressure to match both agent capability and deployment convenience as OpenAI bundles the experience into ChatGPT’s business offerings.
Enterprise buyers will put more weight on how coding agents fit permissions, review practices, and existing developer workflows, not just on code-generation quality.
Third-order effects
If coding agents prove dependable in team settings, software development tools may shift from prompt-based assistance toward managed agents that execute bounded engineering tasks across shared work surfaces.
The earlier Codex API beta that powered Copilot and this rollout suggest model companies are moving up the stack into end-user developer workflows, potentially tightening competition around distribution and enterprise integration.
The trend: This is part of the shift from standalone code-completion models to workflow-embedded agents designed to carry out larger, supervised software-engineering tasks.
wow so far DMs in a shockingly even dead heat between “you made a software engineer and didn't include unlimited use in the $20 plan? fuck you!” and “you made a software engineer and you're not charging $20k a month for it? what the fuck?”
We're releasing Codex, cloud-based software engineering agent that can work on many tasks in parallel. Powered by a new model, codex-1, our best coding model to date and optimized for mergeable code. Feels like a step towards the future of software engineering.
today we are introducing codex. it is a software engineering agent that runs in the cloud and does tasks for you, like writing a new feature of fixing a bug. you can run many tasks in parallel.
it is amazing and exciting how much software one person is going to be able to create with tools like this. “you can just do things” is one of my favorite memes; i didn't think it would apply to AI itself, and its users, in such an important way so soon.
We've been using OpenAI's new coding agent, Codex, internally. tldr; Dispatching agents like this instead of directly guiding them is the future. Quick thread on where it's already added value —and where we're hoping to see it grow. 🧵
“we call this new system, in the grand tradition of OpenAI naming, Codex!” —beautiful, original, definitely not 5 year old product name announced by @gdb this morning
OpenAI just launched Codex, a new AI coding agent inside ChatGPT. It writes, tests, and debugs code in a sandboxed cloud environment, handling tasks asynchronously. This marks a significant step toward autonomous software development. #AI #OpenAI [image]
Will we look back at 2025 as the year engineering teams fundamentally changed how they work? OpenAI just released a research preview of Codex, a cloud-based coding agent. https://techcrunch.com/... [image]
OpenAI showed an environment that they fully configured, such as the Codex-CLI. As configuration, you can also provide steerability and important instructions to the model via MD files. [image]
Introducing Codex — the shiny new AI code editor from OpenAI. It's a fork of Windsurf, which is a fork of VS Code... so basically, it's Microsoft's editor, but we're not on speaking terms, so we took the kids and made our own. [image]
Get ready for a huge wave of posts like “OpenAI Codex is INSANE! Here are 10 examples NO OTHER ai coding assistant can do” followed up by 10 examples of what other ai coding agents can, in fact, already do.
Codex 1 for Pro and Enterprise runs multiple SWE agents on OpenAI compute, connects to your GitHub, shows agents in a terminal sandbox, and auto rewrites code, uses print debugging, and verifies the root cause of issues Greg brockman- “I'm feeling the AGI”
not too impressed by the OpenAI Codex demo. it might be good but the demo doesn't really show that? renaming parameters and extracting constants? they should have autonomously built something live [image]
OpenAI parallel coding Agents are here 🤯 With a custom version of the o3, specialized at coding, Codex can complete multiple tasks at once and even pick up tasks proactively Available on Pro and Enterprise subs 👇 All details: [image]
NEW: OpenAI launched Codex, an AI coding agent that can work on multiple tasks, simultaneously, in the cloud for up to 30 minutes. Agent lead @josh_tobin_ said he wants Codex to eventually complete tasks that take humans “hours or even days,” acting as a virtual teammate. [image]
OpenAI Codex feels insane .. AGI moment - easy and simple to up actual tasks - agents working on their own environment (pretty much like developers working on their own systems) - completes end to end work including tests and PRs sounds simple but the system is insane 100x
🚨 OpenAI just launched Codex, a brand-new autonomous coding agent that can build features and fix bugs on its own. We've been using it @every for a few days, and I'm impressed. I invited Alexander Embiricos (@embiricos), a member of the product staff responsible for Codex, to [vi…
BREAKING: OpenAI announces research preview of Codex in ChatGPT Next-level coding agent within ChatGPT. Pay attention, devs and non-devs! Here is all you need to know: [image]