OpenAI's Persistent Codex Agent: From Chatbot to Always-On Digital Colleague

Nevertheless, OpenAI Persists With New Always-On Agent

While you were sleeping.

OpenAI's Persistent Codex Agent: From Chatbot to Always-On Digital Colleague

Code changes spotted in recent days show OpenAI quietly building a new mode for its Codex agent. The feature, labeled Persistent mode, lets the system keep running on a task until someone explicitly tells it to stop. No more timing out after minutes or hours. The agent stays on the job. Maxwell Zeff first reported the development for WIRED. Engineers reviewed updates to the command-line version of Codex. They found entries describing how the mod…

OpenAI Is Developing a ‘Persistent’ AI Agent

Code reviewed by WIRED reveals the company is developing a feature that enables Codex to continue working proactively until it is “put to sleep.”

OpenAI's Persistent AI Agent: What It Means for You

Photo: Wired OpenAI is building a new kind of AI assistant that will keep working on tasks without you asking it to—and it could change how you use AI tools. Right now, when you ask ChatGPT or similar AI tools to do something, they answer your question and then stop. A new “persistent mode” that OpenAI is testing would flip that. The AI would keep running in the background, spotting things you might need help with and even messaging you with sug…

OpenAI develops persistent AI agent feature for Codex that never stops working

Persistent AI agents could revolutionize software development but pose risks in quality control, necessitating robust oversight mechanisms. The post OpenAI develops persistent AI agent feature for Codex that never stops working appeared first on Crypto Briefing .