Skip to content

ChatGPT Dots, Space and Pages: The New Always-On Workspace, Explained

On 29 September OpenAI introduced dots (always-on agents), Space and Pages. What they are, who can use them, and how to give an agent a goal without losing control.

AI NewsBy Prompt ProfessorOct 16, 20267 min read
A desk with two monitors and a laptop
Photo: Liam Charmer on Unsplash
The short answer

A dot is an agent that keeps working between conversations. Give it a narrow goal, only the apps it needs, and clear rules about what it may do alone.

What a “dot” is

Dots are always-on agents in ChatGPT that take on ongoing work and keep making progress between conversations. You give a dot a goal, connect the apps it needs and define what it can do on its own. Powered by GPT-6 Astra, each dot has its own cloud computer and brings results back for your review.

At launch, dots roll out gradually to eligible Pro and Business Premium users aged 18 and older. Pro access excludes the EEA, Switzerland and the UK at launch, and Enterprise access is a beta that is off by default.

Space and Pages

Space brings Pages, files and related work together, and replaces Library for accounts that have it. Pages are documents you can start from a conversation or a template, edit yourself or with ChatGPT, and share with collaborators who comment using their own ChatGPT. Sharing a Page does not share your private chats or memory.

Give an agent a job description, not a wish

Agents fail in predictable ways: vague goals, too many permissions and no definition of “done”. Treat a dot like a new assistant on day one.

Agent brief template
Goal: [one outcome, measurable]. Context: [who it is for, what good looks like]. You may: [allowed actions]. You must ask before: [sending, buying, deleting, posting]. Check in: [when and how]. Done means: [clear finish line].
AI Agents Guide cover
Recommended for this topic

AI Agents Guide

Understand, build and sell your first AI agent — with no-code and code paths.

$18.90 (was $27) · Look inside

Start with low-risk, repeatable work

  • Weekly research digests on a topic you follow.
  • Tracking changes on a few competitor or supplier pages.
  • Preparing — not sending — follow-up emails after meetings.
  • Organizing files and notes into a Space for a project.

Keep a human in the loop

OpenAI’s own design keeps results coming back for review, and that is the right instinct. Anything public, financial or irreversible should need your approval. If you want to understand how agents use memory, tools and loops — and how people build and sell them — the AI Agents Guide covers it step by step.

Before connecting an app, ask: what is the worst thing this agent could do with this access? If the answer worries you, narrow the permission.

Put it into practice

Open the free Automation Time & ROI Calculator: see what automating a task is really worth. This article is part of The Complete ChatGPT Guide (2026 Edition).

Sources & further reading

Product facts were checked against official sources on 16 October 2026. AI tools change quickly, so check the provider’s current documentation for anything you rely on. Our editorial policy.

Keep reading

Related articles

The weekly AI briefing

Stay ahead of every AI update in five minutes a week.

What changed in ChatGPT, Claude and Gemini, one prompt worth stealing and one tool worth trying.

One email a week. Unsubscribe any time.

We use analytics cookies to understand how visitors use our site. Functional cookies (session and page variant) are always set. See our Privacy Policy.