OpenAI has launched Codex, an AI tool for software engineering tasks, powered by the codex-1 model.

Available as a research preview starting May 16, 2025, Codex runs in a cloud-based, sandboxed environment and integrates with GitHub to access users’ code repositories.

It can write features, fix bugs, answer codebase questions, and run tests. It typically completing tasks within one to 30 minutes.

Codex is initially available to ChatGPT Pro, Enterprise, and Team subscribers, with plans to expand to Plus and Edu users. Users can assign tasks through the ChatGPT interface and monitor progress while continuing their own work.

Codex is designed to reject malicious software requests and runs securely without internet or external API access. OpenAI will introduce rate limits and allow users to purchase additional credits.

🔗 Source: TechCrunch

🧠 Food for thought

1️⃣ The evolution from AI code assistants to autonomous coding agents

OpenAI’s Codex represents a significant shift in the AI coding landscape, moving from simple autocomplete tools to autonomous agents that can tackle entire workflows independently.

Earlier tools like GitHub Copilot functioned primarily as extensions that offered suggestions, while Codex operates as a virtual developer that can write features, fix bugs, and run tests iteratively until achieving success 1.

This evolution reflects the broader industry trend toward AI agents capable of completing end-to-end tasks without constant human guidance, as seen with Anthropic’s Claude and Google’s updated Gemini 2.

The transition has practical implications for development teams. Companies claim approximately 30% of their code is now AI-generated, fundamentally changing how software development work is distributed and executed.

Such autonomous coding capabilities are part of what industry experts in 2025 identify as the shift “from AI tools that respond to prompts to those that can complete entire workflows autonomously” 3.

2️⃣ The intensifying competition in AI coding platforms

OpenAI’s launch of Codex represents a strategic move in an increasingly competitive market where specialized AI coding tools have surged in popularity and investment.

The market has clearly segmented into distinct categories including Web-Based AI Development Tools, IDE Extensions, AI-First IDEs, and CLI Agents, each serving different developer needs and workflows 4.