- OpenAI Dots ChatGPT agent: An always-on AI assistant that keeps working after you log off
- Cloud-powered: Each dot runs its own cloud computer and browser, independent of your laptop
- Plugin ecosystem: Access to over 4,000 plugins for work and personal tasks
- Security controls: Per-app permissions, approval rules, and a transparent activity view
- Eligibility: Rolling out to Pro and Business Premium users first, one dot per account
What Is the OpenAI Dots ChatGPT Agent?
The OpenAI Dots ChatGPT agent is OpenAI's answer to the growing wave of personal AI assistants. Unlike a standard chatbot that waits for your prompts, a dot is an always-on agent: you assign it a project, and it continues working in the background even when your laptop is asleep, locked, or powered off. Under the hood, dots are powered by GPT-6 Astra, and each one runs with its own cloud computer and browser, so nothing depends on your local hardware.
Video Highlights:
- A dot keeps working after you leave — no repeated monitoring or corrections
- Multiple projects can run under one dot without separate chats
- Dots from teammates can collaborate toward a shared goal
- Over 4,000 plugins extend what your dot can access
- Granular security rules control which apps a dot may use
The headline philosophy behind dots is ease of use. OpenAI designed them so you can check progress when you want, but you are not required to babysit the agent. Dots can also interact with local files on your computer, blurring the line between cloud automation and your everyday desktop workflow.
Core Features at a Glance
Dots ship with five standout capability areas. Each one addresses a common frustration with earlier AI assistants that demanded constant supervision.
| Feature | What It Does | Practical Example |
|---|---|---|
| Always-on work | Runs projects in the background on its own cloud computer | Compiling a report overnight while your laptop is off |
| Multi-project handling | Manages several tasks without separate chats | Tracking invoices while planning a trip |
| Plugin access | Connects to 4,000+ plugins for work and life | Noticing a forgotten invoice, preparing it, sending after approval |
| Preference learning | Learns your goals, feedback, and definition of a good result | Adapting report format to your style over time |
| Two-way contact | You can message or call your dot; it can also reach you | Sharing progress or asking for a decision when needed |
Background Execution
- Cloud computer and browser included
- Works while your device is asleep
- Progress check optional
Team Collaboration
- Your dot can work with a teammate's dot
- Dots coordinate toward shared goals
- Built for the future of work
Preference Memory
- Learns from your feedback over time
- Remembers goals and preferences
- Reduces repeated instructions
Give your first dot one well-defined project before layering on multiple tasks. This lets you observe how it interprets instructions and refine its behavior before you trust it with sensitive workflows.
Security and Control Settings
Security is a major focus of the dots launch. Because an always-on agent can act on your behalf across apps, OpenAI built multiple layers of oversight into the system.
| Control Layer | Function | When to Use |
|---|---|---|
| App access rules | Allow, require approval, or block apps | Restrict banking or email access |
| Activity view | See exactly what your dot is doing | Spot unauthorized app attempts |
| Review system | Flags risky or sensitive actions | Prevents compromised accounts |
Before letting a dot touch financial or client-facing tools, review the activity view regularly. The built-in review system catches actions that could compromise your account or expose sensitive information, but your own rules are the first line of defense.
This transparency matters most for small and medium businesses without dedicated IT departments. You can see when a dot attempts to access an app you never authorized, and intervene immediately rather than discovering problems after the fact.
Setup Guide: Getting Your First Dot
Check Plan Eligibility
Dots are rolling out to ChatGPT Pro and Business Premium users. Enterprise, Education, and Healthcare users can join the beta if their workspace administrator enables it.
Wait for Rollout
Access arrives gradually per account. Even Pro users may wait hours or days after the announcement before the feature appears.
Configure App Permissions
Set rules for which apps your dot can access freely, which require approval, and which are blocked entirely.
Assign a Project
Give your dot a clear, bounded first task. Monitor the activity view to confirm it behaves as expected.
Review and Refine
Provide feedback on results so the dot learns your preferences and improves future output.
OpenAI states the first dot comes included with eligible plans, alongside an allowance for deeper work. Multiple dots working together remain a future roadmap item, not a current feature.
Plan Comparison and Launch Checklist
Not every plan gets dots at the same time. Here is how access breaks down based on the announcement:
| Plan Tier | Dot Access | Notes |
|---|---|---|
| ChatGPT Pro | Rolling out | First consumer tier to receive dots |
| Business Premium | Rolling out | Includes one dot with work allowance |
| Enterprise / Education / Healthcare | Beta opt-in | Requires workspace administrator to enable |
| Other plans | Not yet announced | No confirmed timeline |
Before You Use Your First Dot:
- Confirm your plan is Pro or Business Premium
- Define app access rules before assigning tasks
- Start with one low-risk project
- Review the activity view daily in week one
- Give explicit feedback so the dot learns preferences
For now, each account supports a single dot. OpenAI has signaled that multiple cooperating dots are on the roadmap, but there is no confirmed release date yet.
FAQ
Q: Does the OpenAI Dots ChatGPT agent work when my computer is off?
Yes. Each dot runs on its own cloud computer and browser, so it continues working whether your laptop is asleep, locked, or powered off. You can check progress anytime, but constant monitoring is not required.
Q: Which plans include access to dots?
Dots are rolling out to ChatGPT Pro and Business Premium users. Enterprise, Education, and Healthcare workspaces can join a beta if their administrator enables it. The first dot is included with eligible plans.
Q: Can I run multiple dots at once?
Currently, accounts are limited to one dot. OpenAI has described multi-dot collaboration, including dots working with teammates' dots, as a future direction rather than a feature available today.
Q: How secure are dots?
You control which apps your dot can access, with options to allow, require approval, or block. An activity view shows what the dot is doing, and a review system flags actions that could compromise your account or expose sensitive data.