OpenAI Is Developing a 'Persistent' AI Agent
Wired reports that OpenAI is testing a new "Persistent mode" for Codex that would let the agent keep working until explicitly "put to sleep," proactively creating follow-up tasks for itself across sessions and using prior interactions and knowledge of the user to decide what to do next. Code alluding to the feature was spotted in the product's code base, though it has not been rolled out or…
OpenAI is in the midst of developing a Persistent mode for its Codex AI agent, according to reports from Wired. This new feature would allow the agent to continue working on tasks even after they are initially completed, making proactive decisions on what to do next based on previous interactions and knowledge of the user. The persistent mode appears in Codex's reasoning effort menu, where users can set the level of computing resources the AI can use.
When activated, the agent will keep working until explicitly put to sleep, marking a significant departure from currently available modes that stop after a few minutes or hours. A separate file in the code base introduces the concept of proactivity within the Persistent mode. This feature would enable the AI to create follow-up tasks for itself even after completing an initial request, and continue working on these tasks across sessions.
The agent is also designed to communicate with the user sparingly, without being prompted, but must seek approval before altering anything outside the user's system. These instructions are intended to ensure the AI remains safe and controlled. The proactivity feature seems to be more than just a command line tool, suggesting it could be integrated into other platforms as well.
Written by urgent.news from Slashdot's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
