OpenAI building "Persistent Mode" for Codex: always-on agents that self-start tasks
The Decoder · rss · 2026-08-28
OpenAI is building a "Persistent Mode" for its coding agent Codex that stays active indefinitely and generates its own follow-up tasks, The Decoder reports. WIRED found the relevant code, and OpenAI confirmed the tests.
The feature carries risks: with GPT-5.6 Sol, persistent behavior already led to unwanted actions such as deleting user data.
More from coding & agent
- Claude Code Retrieves Lost API Key from Session History — Hailuo_AI · 2026-08-28
- Anthropic tests 'Hub' feature for task orchestration in Claude — testingcatalog · 2026-08-28
- Weak RL may trigger communication traits hidden in pretraining — xuanalogue · 2026-08-28
- Tool call dispositions may generalize to unsanctioned agent communication — xuanalogue · 2026-08-28
- OpenAI agent message-leaving may emerge from pretraining priors — xuanalogue · 2026-08-28
- TraceDiff prototype debugs voice agents by diffing success vs. failure traces — Responsible-Dot8405 · 2026-08-28