OpenAI Ships Workspace Agents: Always-On AI Teammates That Live Inside Your Company's Tools
OpenAI launched workspace agents for ChatGPT business plans — persistent, shared agents that integrate with Slack, Linear, and other tools to run complex workflows autonomously in the background. It's the company's most aggressive enterprise play yet, and it landed the same week Google unveiled its own agent platform.
OpenAI released workspace agents for ChatGPT on Wednesday, introducing persistent AI agents that can be built once, shared across teams, and left to run complex workflows in the background — a significant architectural shift from the prompt-and-response paradigm that has defined ChatGPT since launch. As @OpenAI put it in its announcement: "Shared agents that can handle complex tasks and long-running workflows across tools and teams."
The feature is available as a research preview for ChatGPT Business, Enterprise, Edu, and Teachers plans. Greg Brockman, OpenAI's president, described the technical underpinning: agents run on a "cloud-hosted Codex harness" and can be connected to external tools, given recurring tasks, and invoked from surfaces like Slack. As @gdb wrote: "Hook them up to tools, give them recurring tasks, and talk to them from surfaces like Slack." The implication is clear — these aren't chatbots you query; they're background processes you deploy.
Get our free daily newsletter
Get this article free — plus the lead story every day — delivered to your inbox.
Want every article and the full archive? Upgrade anytime.
No spam. Unsubscribe anytime.