Posts
Introducing EVMbench
OpenAI and Paradigm introduce EVMbench, a benchmark evaluating AI agents’ abi...
Harness engineering: leveraging Codex in an agent-fi...
By Ryan Lopopolo, Member of the Technical Staff
Sea's View on the Future of Agentic Software Develop...
Sea Limited's CPO explains why the company is deploying Codex across engineer...
An open-source spec for orchestration: Symphony
Learn how Symphony, an open-source spec for Codex orchestration, turns issue ...
Introducing the Codex app
Introducing the Codex app for macOS—a command center for AI coding and softwa...
Speeding up agentic workflows with WebSockets in the...
A deep dive into the Codex agent loop, showing how WebSockets and connection-...
Helping developers build safer AI experiences for teens
OpenAI releases prompt-based teen safety policies for developers using gpt-os...
Codex Security: now in research preview
Codex Security is an AI application security agent that analyzes project cont...
OpenAI Codex and Figma launch seamless code-to-desig...
OpenAI and Figma launch a new Codex integration that connects code and design...
Unlocking the Codex harness: how we built the App Se...
Learn how to embed the Codex agent using the Codex App Server, a bidirectiona...
OpenAI Launches GPT-5.5 Biological Safety Bug Bounty...
OpenAI has launched a specialized bug bounty program targeting GPT-5.5 to ide...
OpenAI Acquires Promptfoo to Standardize AI Model Ev...
OpenAI plans to acquire Promptfoo to strengthen its evaluation infrastructure...
OpenAI Expands API Capabilities With New Computer En...
OpenAI has integrated a sandboxed computer environment into its Responses API...
Introducing Advanced Account Security
Introducing Advanced Account Security: phishing-resistant login, stronger rec...