OpenAI has announced its commitment to publish results from its AI safety tests more frequently, aiming to enhance transparency and trust in its AI systems. The move is part of a broader initiative to prioritize safety and accountability in artificial intelligence development.

OpenAI has launched Codex, a coding tool designed for developers to delegate simple programming tasks to an AI agent that generates production-ready code. Accessible via the ChatGPT web app, Codex operates within a user’s development environment and can be customized through an “AGENTS.md” file for improved effectiveness. Built on a fine-tuned model, Codex leverages reinforcement learning to enhance its coding capabilities.