- OpenAI Dots is an always-on ChatGPT assistant powered by GPT-6 Astra, announced at DevDay on September 29, 2026.
- Each Dot runs on a dedicated cloud computer with its own browser for long-running tasks.
- Dots connect to 4,000+ apps and follow up through ChatGPT, email, Slack, and SMS.
- Use this guide to go from creation to scheduled automation in six steps.
- Access depends on your ChatGPT plan, workspace settings, and rollout stage.
What the OpenAI Dots ChatGPT Assistant Is
The OpenAI Dots ChatGPT assistant is a new kind of AI helper: a persistent agent that keeps working toward a goal instead of stopping when a chat ends. OpenAI introduced Dots at DevDay on September 29, 2026, and every Dot is powered by GPT-6 Astra, the flagship model built for complex reasoning, computer use, research, and end-to-end work.
Unlike a normal ChatGPT conversation, a Dot maintains context across sessions. It can run multi-step tasks on its own cloud computer, check in with you through connected channels, and continue scheduled work between conversations. OpenAI's official safety documentation now includes a dedicated Dots section, reflecting how deeply the product is integrated into the ChatGPT agent ecosystem.
| Feature | Normal ChatGPT Chat | OpenAI Dot |
|---|---|---|
| Session model | Single conversation | Persistent goal across sessions |
| Execution | Responds when prompted | Works between conversations |
| Environment | No dedicated machine | Own cloud computer + browser |
| Model | Varies by plan | GPT-6 Astra |
| Follow-up | You re-prompt | Proactive updates via ChatGPT, email, Slack, SMS |
Think of a Dot as a delegated coworker, not a chat window. You define the goal once, and the Dot organizes its own work sessions, tools, and follow-ups around that goal.
How to Set Up Your First Dot
Creating a Dot takes only a few minutes if you prepare a clear goal and boundaries in advance. Dots are created from ChatGPT on desktop web or the supported desktop app, with more advanced configuration centered on desktop experiences.
Start a New Dot
Open the Dots experience inside ChatGPT on desktop web or the desktop app and choose to create a new dot.
Choose a Name
Pick a short, descriptive name based on responsibility, such as a project, workflow, research topic, or recurring operational task.
Define Its Goal
Write a clear ongoing outcome the Dot should continually work toward, not a single isolated prompt.
Configure Autonomy
Set instructions and boundaries that determine how independently the Dot acts and when it should ask for approval.
Attach Context and Files
Add relevant files and supporting information so the Dot has the project context it needs.
Open the Activity Dashboard
Use the activity view to review completed actions, follow current tasks, and manage ongoing configuration.
A goal like "help with marketing" gives the Dot nothing stable to organize around. Specify the outcome, cadence, and boundaries, for example: "Compile a weekly competitor summary every Friday and flag pricing changes over 10%."
Cloud Computer and Connected Apps
Two capabilities separate Dots most sharply from ordinary ChatGPT use: the dedicated cloud computer and the connected-app system.
The Dedicated Cloud Computer
Every Dot has its own cloud-based computer with a browser. Work there can continue while you move on to other things, which makes the environment ideal for multi-step workflows like research, form-filling, and site monitoring. You can open the computer view to inspect what the Dot is doing and step into the session when manual input is useful.
The cloud computer is separate from optional local-computer access. Cloud execution happens in the Dot's own remote environment; local access involves your device and is an optional, separately controlled capability.
Connected Apps and Channels
Connected apps expand what a Dot can reach, and each connection follows the permissions granted by the user or workspace. The launch materials highlight 4,000+ app connections as a core selling point.
| Connection | Category | What It Enables |
|---|---|---|
| Slack | Team communication | Work with permitted workspace channels and context |
| Communication | Handle approved messages and follow-ups | |
| Files | Knowledge and context | Use project documents and reference material |
| Workspace apps | Productivity | Extend workflows with external information and actions |
| Channels (SMS, etc.) | Notifications | Exchange updates and task context outside ChatGPT |
Scheduled Tasks and Automation
Scheduled work turns a Dot into an ongoing assistant rather than a one-time chat. Instead of prompting the agent each morning, you assign work that runs later or repeats on a schedule.
| Task Type | Example | Best For |
|---|---|---|
| Reminders | Surface a deadline three days ahead | Time-sensitive follow-ups |
| Recurring checks | Review a competitor site each Monday | Ongoing monitoring |
| Proactive updates | Send a digest after each work cycle | Staying informed passively |
| Recurring summaries | Weekly calendar and inbox recap | Executive-style briefings |
Create a Scheduled Task
Tell the Dot what should happen and when, either at a specific future time or on a recurring schedule.
Let Recurring Checks Run
The Dot reviews updates, checks connected information, or prepares recurring summaries on its own.
Receive Proactive Updates
Results arrive through your configured channels without starting a new conversation.
Review Task Activity
Open the task management view to see active, scheduled, and completed work over time.
The strongest use cases are workflows where context must persist: ongoing research that evolves, recurring monitoring of changing information, and multi-step agentic work spanning research, browser interaction, and follow-up actions.
Permissions, Plans, and Availability
Permissions define the boundaries within which a Dot operates. Both individual users and workspace administrators have controls.
| Control | Scope | What It Governs |
|---|---|---|
| Connected app permissions | User and workspace | Which apps and services the Dot may access |
| Cloud computer access | Dot capability | Whether the Dot can run browser-based computer tasks |
| Local computer access | Optional user capability | Access involving the user's own device |
| Autonomy settings | User configuration | How independently the Dot continues work |
| Workspace availability | Administrator | Whether members can use Dots at all |
| Enterprise controls | Enterprise admin | Workspace-level restrictions and governance |
Availability depends on your ChatGPT plan, workspace configuration, rollout status, and platform. Desktop web and the desktop app are the primary surfaces for creating and managing Dots; mobile access varies by account and rollout stage, and Enterprise workspaces receive Dots through a controlled, admin-managed deployment.
Because rollout is staged, confirm Dots appear in your ChatGPT interface before building workflows around them. Workspace members should verify that administrators have not disabled Dots for the organization.
Practical Use Cases and Launch Checklist
Ongoing Research
- Assign an evolving topic
- Dot collects and organizes findings
- Continues across work sessions
Calendar and Inbox
- Review upcoming events
- Prepare recurring summaries
- Flag important meetings
Recurring Monitoring
- Watch changing information
- Surface meaningful updates
- Reduce manual checking
Getting Started With OpenAI Dots:
- Confirm Dots are available on your ChatGPT plan
- Create a Dot with a clear name and goal
- Configure autonomy boundaries and approval rules
- Connect one app such as Slack or email
- Assign a first scheduled task
- Review the activity dashboard after the first cycle
Delegate one narrow recurring task first. Watching how the Dot handles a single weekly report teaches you its reporting style and autonomy behavior before you hand over anything critical.
FAQ
Q: What is the OpenAI Dots ChatGPT assistant?
Dots are always-on AI agents inside ChatGPT, announced at DevDay on September 29, 2026. Powered by GPT-6 Astra, each Dot has its own cloud computer and browser, connects to apps, and keeps working on long-running tasks between conversations.
Q: Which model powers OpenAI Dots?
GPT-6 Astra powers every Dot. OpenAI positions Astra as its flagship model for complex reasoning, coding, computer use, research, and end-to-end agentic work.
Q: Can I use Dots on mobile?
Mobile access varies by account, platform, and rollout stage. Creating and configuring Dots is centered on desktop web and the ChatGPT desktop app.
Q: How do I control what a Dot can access?
Permissions cover connected apps, cloud computer use, optional local computer access, and autonomy settings. Workspace and Enterprise administrators can further restrict or disable Dots for members.