OpenAI launches Codex, a cloud-based software engineering agent that automates more of the work developers do in a safer and less opaque way.
OpenAI Launches an Agentic, Web-Based Coding Tool
Automating Software Development with Safety and Explainability
OpenAI is launching a cloud-based software engineering agent called Codex, which aims to help advanced developers automate more of their work in a safer and less opaque way than existing tools. This tool is designed for professional coders rather than amateur vibe coders.
OpenAI is a research organization focused on developing and advancing artificial intelligence (AI) in a way that benefits humanity.
Founded in 2015, OpenAI has made significant contributions to natural language processing, computer vision, and robotics.
Their AI models, such as GPT-3, demonstrate impressive capabilities in generating human-like text and responding to user queries.
With the goal of creating AI that is safe and beneficial for society, OpenAI collaborates with leading researchers and institutions worldwide.
Codex can generate lines of code and move through directories and run commands inside a virtual computer, automating more of the work that developers go through when writing code. According to Alexander Embiricos, a member of the product team at OpenAI working on agents, ‘We’re about to undergo a pretty seismic shift in terms of how developers can be most accelerated by agents.‘
A codex is an ancient term for a book, typically made of multiple pages bound together.
The word 'codex' comes from the Latin 'caudex', meaning 'trunk of a tree.' In ancient Rome, codices were often written on parchment or papyrus and contained important documents, literary works, and historical records.
The codex format eventually replaced the earlier scroll format, becoming the standard form for books in the Western world.
Today, the term 'codex' is used primarily by historians and scholars to describe early manuscripts.

A Safer and More Explainable Approach
One key challenge with vibe coding is that delegating to AI can result in software that is opaque and more difficult for a person to understand and fix when bugs creep in. OpenAI says the model behind Codex has been trained to explain what it is doing more clearly and help developers fix what they are building.
The use of a virtual computer makes the system safer by design, according to Embiricos. ‘That’s really the way that we think most development is going to happen in the future, he says. The agent will work on its own computer and will delegate to it.‘
A virtual computer, also known as a virtual machine (VM), is a software emulation of a physical computer.
It runs its own operating system and can execute programs independently from the host computer's OS.
Virtual computers are created using hypervisor software, which allocates resources such as CPU, memory, and storage to the VM.
This allows for multiple VMs to run on a single physical machine, increasing efficiency and flexibility.
Industry Adoption and Future Plans
Codex is already being used by outside companies including Cisco, Temporal, Superhuman, and Kodiak. OpenAI has launched two other agentic AI tools over the past year: Operator, which controls a web browser and can automate online chores, and Deep Research, which carries out detailed web search and analysis in order to compile reports.
Josh Tobin, who leads the agents research team at OpenAI, says Codex reflects a bigger vision for ChatGPT to evolve from a chatbot into a teammate. ‘We think that ChatGPT will become almost like a virtual coworker, he says. Where you can go to it not just for answers to quick questions [but also to] collaborate with it on larger chunks of work across a wide range of different tasks.‘