Back to AI updates
OpenAINew features

Introducing dots

This is for anyone who wants ChatGPT to keep working on a task in the background instead of only responding while you're actively chatting, useful for ongoing research, monitoring or multi-step projects. It's currently limited to eligible Pro and Business Premium subscribers, so most free or basic-plan users won't see it yet.

Read the original source 3 min read

What changed

OpenAI launched 'dots', always-on AI agents powered by GPT-6 Astra that carry out ongoing work between conversations, rolling out gradually to eligible ChatGPT Pro and Business Premium users.

Why it matters to your work

This is for anyone who wants ChatGPT to keep working on a task in the background instead of only responding while you're actively chatting, useful for ongoing research, monitoring or multi-step projects. It's currently limited to eligible Pro and Business Premium subscribers, so most free or basic-plan users won't see it yet.

When this could help

Useful when you have ongoing tasks you'd delegate to a capable assistant who can work independently, check back in, and take actions across your connected tools.

Dots are persistent, always-on agents with their own cloud computer, able to work across Slack, Teams, email (texting coming soon), with optional laptop access, proactive research, and safety monitoring — a step beyond one-off chat toward a delegated ongoing assistant.

How to get started

  1. Each dot has its own cloud computer and can optionally be granted permission to connect to your laptop for local file access.

    Provider guidance ↗
  2. Configure Custom Rules to allow, require approval for, or block specific actions your dot can take.

    Provider guidance ↗
  3. Use Activity View to review what your dot has done, including screenshots and action logs.

    Provider guidance ↗
  4. Communicate with your dot via Slack, Teams, or the ChatGPT app (texting coming soon); dots can also do proactive research using read-only tools.

    Provider guidance ↗

Give AI a clearer brief

Situation
You have an ongoing task or set of responsibilities you'd like to delegate to a persistent assistant that can work across your tools.
Consultant
Act as a capable, always-available assistant.
Action
I'd like you to [describe the ongoing responsibility, e.g. 'monitor my inbox for customer escalations and draft responses for my review']. Use Custom Rules: [list what to allow, what needs approval, what to block]. [Fill in your real task and rules.]
Language
Brief status updates and flagged items requiring my decision.
Ask
Before starting, confirm which tools and accounts you'll need access to, and walk me through the Custom Rules you recommend for this task.

SCALA suggested prompt · Adapt the context and check the result.

Learn the SCALA method →

Build it into your routine

  1. Delegated ongoing work

    Identify a recurring responsibility (monitoring, research, drafting) and set it up as a dot task with appropriate Custom Rules, then review Activity View periodically to check its work and adjust rules as needed.

What to check

  • Conversations with a dot don't count toward usage limits, but tasks in Codex and Work do. Source ↗
  • OpenAI's safety monitoring can pause or stop a dot if it detects a safety concern. Source ↗
  • Dots can still make mistakes — review Activity View and set Custom Rules accordingly. Source ↗

Sources & provider guides

Official documentation checked 12 Oct 2026. SCALA prompts and workflows are suggestions, not claims of hands-on testing.

Source & availability

This is a summary of a published provider announcement, with practical context from SCALA’s feed.

Plan, region, workspace and rollout conditions may affect access in Australia. Check the original source for current availability.

openai.com