Skip to content
  • NVDA
  • AAPL
  • MSFT
  • AMD
  • TSLA

OpenAI Developing Persistent AI Agent

Code review reveals proactive agent feature

By TMRO Staff·1 min read
OpenAI Developing Persistent AI Agent
Wired

Key points

  • WIRED reviewed code showing OpenAI developing persistent agent feature.
  • Feature enables Codex to continue working proactively.
  • Agent works until it is 'put to sleep.'
  • Details from WIRED article published August 27, 2026.

What happened

WIRED reported on August 27, 2026, that code reviewed by the publication reveals OpenAI is developing a feature that enables Codex, its AI coding tool, to continue working proactively until it is "put to sleep." The report, based on an examination of code, indicates the company is building a persistent agent capability within Codex.

Why it matters

This development signals a shift toward more autonomous AI agents that can operate without constant user input. A persistent agent could handle longer tasks and work independently, potentially changing how developers interact with AI tools. However, details about the feature's functionality, availability, or safety measures were not disclosed in the report.

Why it matters

This development signals a shift toward more autonomous AI agents that can operate without constant user input, potentially changing how developers interact with AI tools.

The data

TMRO coverage, last 90 days

OpenAI
10 storieslast on 29 Aug 2026

TMRO Report archive ·

Sources

TMRO Report writes original coverage based on the material listed above.