OpenAI Codex

OpenAI has announced a focused 28-day sprint to improve Codex, its coding assistant. The company says the sprint is aimed at addressing user feedback and shipping rapid fixes and refinements to the tool. Developers following the project are watching to see how quickly improvements land and whether the intensive schedule leads to noticeable gains in reliability and performance.

A developer has released a command-line tool that compresses tokens before they are sent to AI coding assistants such as OpenAI's Codex, aiming to reduce API bills for heavy users. The tool compresses prompt and context tokens to lower costs while attempting to preserve output quality. Early commenters are discussing how much it can realistically save and whether compression risks degrading results.

A demonstration shows OpenAI's ChatGPT-6, codenamed Astra, playing World of Warcraft with no visual input and clearing the orc starting zone in 40 minutes without a single death. Built on a private server via a single prompt in OpenAI's Codex, the model created its own pathfinding system to navigate the game. Observers are debating what this means for AI agents tackling complex, unfamiliar environments.

OpenAI has made its automatic code review feature free for all Codex users. The tool reviews code changes automatically, giving developers feedback without a paid tier. The move is being discussed as a way to pull more developers into OpenAI's coding ecosystem at a time when automated code review and AI coding assistants are a highly competitive market.

OpenAI says it will ship daily improvements to its Codex coding agent, a pledge made as users increasingly compare it with Anthropic's Claude Opus 5.5. Developers on social media are debating which model handles real-world coding tasks better, with many reporting a shift toward Claude for complex work. The exchange highlights intensifying competition in the AI coding-assistant market.

Software developers are weighing OpenAI's Codex app against Anthropic's Claude Code, comparing how the two AI coding assistants handle real programming tasks. Discussions focus on differences in speed, code quality, pricing and workflow integration, with opinions split on which tool performs better for day-to-day development work.

Software developers are increasingly choosing Anthropic's Claude Code as their AI coding assistant over OpenAI's Codex. The shift suggests Claude Code's performance and workflow integration are winning over programming teams, intensifying competition in the fast-growing market for AI developer tools between the two leading AI labs.

OpenAI has announced a text watermarking system designed to make AI-generated content from ChatGPT, Codex, and its API identifiable. The move is a compliance step under the EU AI Act, which requires providers of generative AI systems to mark machine-generated text. The announcement is drawing attention from developers and regulators tracking how AI companies implement transparency requirements.

On October 5, 2026, OpenAI announced that ChatGPT and Codex will begin embedding invisible watermarks into text output for EU users. The feature, called textGrain, is being linked to compliance with the EU AI Act, which requires disclosure of AI-generated content. Developers building on OpenAI's models are watching closely to see how the change affects downstream applications.

OpenAI has introduced an invisible text watermark to ChatGPT and its Codex coding tool, designed to flag content as AI-generated. The system comes with stated limitations, and OpenAI has not fully detailed how reliably it can detect watermarked text or under what conditions it may be removed, leaving questions about its practical impact on detecting AI-written material.

OpenAI has released a Codex app, placing AI coding agents at the center of its developer tooling. The launch underscores a broader industry shift from autocomplete-style assistants to autonomous agents that can write, test and manage code with limited supervision. Developers are debating how much real work these agents can handle and what the change means for software engineering jobs and workflows.

OpenAI is drawing attention to Codex Originals, a name associated with its Codex software line. Details about what Codex Originals actually covers remain scarce, as no additional information or figures have accompanied the mentions. Observers in the software community are watching closely, given OpenAI's prominence in developer tools and AI coding assistants, and further announcements may follow.

A side-by-side comparison put three AI coding tools — Anthropic's Claude Code, OpenAI's Codex and Google's Antigravity — to the test, asking each to build the same Android app in under 30 minutes. The writeup compares results and names which assistant the author would use again, adding to ongoing debate over which AI coding agent is most practical.

OpenAI is introducing invisible watermarks into text produced by ChatGPT and Codex for users in the European Union. The measure is aimed at identifying AI-generated content, amid ongoing regulatory pressure in Europe for transparency around synthetic text and machine-generated output.

OpenAI's text watermarking system detects about 80% of 200-token passages at a 1% false-positive target, but accuracy falls to 17% once 25% of the words are changed. The tool is available to API users on an opt-in basis, and text from ChatGPT and Codex will be marked in the EU within weeks. Observers are highlighting the system's vulnerability to simple paraphrasing and edits.

OpenAI is watermarking text generated by ChatGPT and Codex for users in the European Union, using a method referred to as textGrain. The move is being discussed as a step toward making AI-generated content identifiable, in line with EU transparency expectations around artificial intelligence. Reaction so far appears focused on what this means for users and downstream AI companies.

OpenAI's Codex, the company's AI coding agent built on its GPT models, is generating renewed discussion as developers and tech commentators weigh it against ChatGPT. The conversation centres on how the two tools fit together: ChatGPT as the general assistant and Codex as a specialised tool for writing and reviewing code inside developers' workflows.

Heise is hosting a live webinar on developing, deploying and reviewing software with AI coding agents in Xcode. The session covers Claude Code, GitHub Copilot and OpenAI Codex, with a focus on using the tools without compromising security or data privacy. The announcement is circulating among developers interested in AI-assisted programming workflows.

OpenAI has begun embedding invisible watermarks into text generated by ChatGPT and its Codex coding tool for users in the European Union. The move, reported by BleepingComputer, is being read as a step toward compliance with EU transparency rules around AI-generated content, allowing machine-generated text to be identified. The news is drawing attention to how AI firms verify and label output, and to questions about privacy and how such detection will work in practice.

Posts circulating online claim that the system prompt for OpenAI's rumoured GPT-6 model has leaked, describing a setup referred to as the 'Sol Codex' that allegedly uses advanced systems engineering and autonomy features. The claim has not been verified by OpenAI, and no official confirmation of the model or the leak exists. Tech and security communities are debating whether the document is genuine or fabricated.

A blind developer shared findings on an accessibility bug in OpenAI's Codex when used with the NVDA screen reader, where answers appear cut off or truncated as if only the visible screen content is read before jumping back to the prompt. The post invites other blind users experiencing the same problem to compare notes and solutions.

OpenAI has launched GPT-6.1 Sol, a model said to deliver Astra-like performance at a lower cost. Early access begins through ChatGPT Work and Codex, while broader chat availability will have to wait. The rollout is drawing attention as a bid to bring high-end AI capability to budget-conscious users and developers.

OpenAI's Codex Cloud is in the spotlight as a cloud-based coding agent that can work on software tasks in parallel from OpenAI's servers. Developers are discussing how it integrates with their workflows, its usefulness for automating coding work, and how it compares with rival AI coding tools.

OpenAI has updated its Codex coding agent with reusable cloud environments that follow developers across devices, according to SiliconANGLE. The feature lets workspaces persist between sessions and machines, so developers can pick up projects without reconfiguring setups. It is being discussed as part of OpenAI's broader push to make its developer tools more practical for ongoing software work.

OpenAI has updated its Codex coding agent so developers can create reusable cloud environments that persist across devices. The change means work set up in Codex can be picked up later or on another machine without reconfiguring. The update is aimed at making the agent more practical for ongoing software projects.

OpenAI's Codex Security Cloud reviews GitHub repositories and new commits in the cloud, investigates suspected security flaws, and drafts fixes for human approval. Because the scanning runs continuously in the cloud, analysis continues even after a developer closes their laptop. The launch raises questions about where automated commit analysis fits in existing code review workflows.

OpenAI's Codex pricing page indicates that Pro subscribers get unlimited usage of GPT-5.6 within Codex, OpenAI's coding agent. The claim is drawing attention among developers comparing plans and weighing what unlimited agent usage means for coding workflows and the value of the Pro subscription.

OpenAI introduced a new ChatGPT feature called 'Dots' at its DevDay 2026 event, described as an agent capable of working around the clock. The announcement is part of a broader push into agentic AI, alongside tools such as Codex and GPT-based agents. The full details and capabilities of Dots have not been independently confirmed.