OpenAI team shares how it runs and governs fleets of persistent agents internally
romainhuet · x · 2026-11-13
An OpenAI team (steipete et al.) has published their practices for running persistent agents inside the company.
Key thesis: persistent agents are more like contractors than traditional SaaS — they act proactively in the background without human intervention and can fulfill entire job functions, but can also take misaligned actions and write arbitrary code with unintended consequences, posing a major challenge for enterprise IT.
The post details how OpenAI manages its internal fleet of agents (like Openclaw and Dots): permissioning, behavioral boundaries, and operational practices. A rare first-hand account of agent governance at a frontier lab.
Related event: OpenAI Shares Best Practices for Managing Persistent Agents at Scale(2 posts)→
More from coding & agent
- Gemini 4 Hydrogen builds a 3D desert ranch in 56s vs 6+ minutes for 3.8 Flash — minchoi · 2026-11-13
- Interactive 3D dining room with working lights and openable fridge built in 30 minutes — minchoi · 2026-11-13
- Carbon orchestrates a swarm of Hydrogen workers to build a crimson moth in 45 minutes — minchoi · 2026-11-13
- Gemini 4 Hydrogen turns Google Sheets into a living landscape, cell by cell — minchoi · 2026-11-13
- Streaming output and visible progress, not faster models, keep users from abandoning slow AI features — XxFreakhawkxX · 2026-11-13
- Cloudflare ships 4 models in 3 weeks as it rebrands its AI team a 'neo lab' — thursdai_pod · 2026-11-13