- OpenAI Dots vs ChatGPT: Dots are always-on agents; ChatGPT chats are session-based conversations.
- Core engine: Every Dot runs on GPT-6 Astra, OpenAI's flagship model for complex reasoning.
- Cloud computer: Each Dot gets its own cloud PC and browser that keep working independently.
- Connectivity: Dots connect to 4,000+ apps plus Slack, email, and ChatGPT channels for proactive follow-ups.
- Best use: Choose ChatGPT for quick Q&A; choose a Dot for long-running, multi-step work.
OpenAI Dots vs ChatGPT: The Core Difference
OpenAI Dots, announced at DevDay on September 29, 2026, represent a fundamentally different way of working with AI than a standard ChatGPT conversation. A ChatGPT chat responds to the prompt you type, then waits. A Dot is an always-on AI agent powered by GPT-6 Astra that keeps working toward an assigned goal between conversations, following up through ChatGPT, SMS, email, and Slack.
| Aspect | ChatGPT (Standard Chat) | OpenAI Dots |
|---|---|---|
| Runtime model | Session-based, waits for prompts | Always-on, persistent goals |
| Underlying model | Varies by plan | GPT-6 Astra flagship model |
| Computer access | Uses your current session only | Dedicated cloud computer + browser |
| App connections | Limited within chat context | 4,000+ connected apps |
| Follow-up channels | Inside the chat window | ChatGPT, SMS, email, Slack |
| Task duration | Single conversation | Long-running, multi-session work |
| Scheduling | Manual prompting | Scheduled and recurring tasks |
The simplest way to frame it: ChatGPT answers, a Dot executes. If your task ends when you close the tab, use ChatGPT. If the work should continue after you walk away, use a Dot.
Architecture: How a Dot Works Behind the Scenes
Three architectural features separate Dots from ordinary ChatGPT usage. Together, they turn a conversational model into a persistent worker.
Dedicated Cloud Computer
- Own remote PC per Dot
- Runs a full browser
- Works while you're offline
- Independent of your local machine
GPT-6 Astra Engine
- Built for complex reasoning
- Strong computer use abilities
- Handles research and coding
- Designed for end-to-end work
Connected Tools
- 4,000+ app connections
- Slack, email, messaging channels
- Acts inside permitted services
- Follows user and workspace permissions
This is the single biggest functional difference from regular ChatGPT. A Dot's cloud computer continues multi-step browser work—research, form filling, monitoring—without keeping your browser tab or laptop open. You can open the Dot's computer view anytime to inspect its state or intervene manually.
OpenAI's official safety documentation already includes a dedicated Dots section, reflecting that this architecture introduces new permission boundaries: connected-app access, cloud versus local computer use, and administrator-level workspace controls all exist specifically because Dots act with more independence than a chat window.
When to Use Which: Task-by-Task Comparison
Match the tool to the job. The table below maps common workflows to the better option.
| Workflow | Best Choice | Why |
|---|---|---|
| Quick factual question | ChatGPT | One prompt, one answer |
| Drafting an email | ChatGPT | Single-session writing task |
| Ongoing research topic | Dot | Evolves across multiple work sessions |
| Recurring monitoring | Dot | Scheduled checks without new prompts |
| Calendar review prep | Dot | Uses connected calendar, repeats weekly |
| Follow-up reminders | Dot | Surfaces tasks at the right time |
| Cross-app workflows | Dot | Spans Slack, email, and connected apps |
| One-off code snippet | ChatGPT | Fast, contained output |
If you find yourself re-explaining the same context to ChatGPT every week, that workflow is a strong candidate for a Dot. Dots persist context, instructions, and goals so you assign the outcome once.
Setting Up Your First Dot
If the comparison convinces you to try a Dot, the setup flow is straightforward. Dots are created from ChatGPT on desktop web or the supported desktop experience, with advanced configuration centered on desktop.
Create Your Dot
Open the Dots experience in ChatGPT and start a new dot. Give it a clear, descriptive name tied to the project or workflow it will own.
Define the Goal
Write the ongoing outcome the Dot should work toward—not a single prompt. A focused goal gives the Dot stable direction across sessions.
Set Working Instructions and Autonomy
Describe priorities, boundaries, and how independently the Dot should operate, including when it should ask for approval before acting.
Connect Apps and Attach Context
Authorize the connected apps and communication channels the Dot needs, and attach files that provide project context.
Assign the First Task and Review Activity
Give the Dot a concrete task linked to its goal, then monitor progress through its activity dashboard and task management view.
Dots access is plan-dependent and tied to the current rollout. Workspace administrators control whether members can use Dots, and Enterprise deployments apply their own governance. Verify your plan supports Dots before building workflows around them.
Permissions, Safety, and Controls
Because Dots act autonomously, OpenAI layered permissions more deeply than in standard ChatGPT. Every capability boundary is configurable.
| Control | Scope | What It Governs |
|---|---|---|
| Connected app permissions | User and workspace | Which apps a Dot may access |
| Cloud computer access | Dot capability | Browser-based, multi-step tasks |
| Local computer access | Optional user capability | Access to your own device, separate from cloud |
| Autonomy settings | User configuration | How independently the Dot acts between interactions |
| Workspace availability | Administrator | Whether members can use Dots at all |
| Enterprise access controls | Enterprise admin | Organization-wide restrictions and governance |
Before Giving a Dot Broad Access:
- Review every connected-app permission grant
- Set autonomy boundaries and approval rules
- Keep cloud and local computer access separate
- Confirm workspace admin policies allow Dots
- Check the activity dashboard regularly for unexpected actions
Begin with a single-purpose Dot—one research topic or one recurring report—on minimal permissions. Once its behavior matches expectations, expand its connected apps and autonomy gradually.
FAQ
Q: Is OpenAI Dots a replacement for ChatGPT?
No. Dots live inside the ChatGPT ecosystem and complement standard chats. Use ChatGPT for quick questions and single-session tasks; use a Dot for long-running, multi-step work that benefits from an always-on agent with its own cloud computer.
Q: What model powers OpenAI Dots?
Dots are powered by GPT-6 Astra, OpenAI's flagship model designed for complex reasoning, computer use, research, and end-to-end agentic work.
Q: How do Dots follow up with me outside ChatGPT?
Dots can communicate through ChatGPT as well as connected channels including SMS, email, and Slack, delivering proactive updates when scheduled or long-running tasks complete.
Q: Do Dots run on my own computer?
Each Dot uses a dedicated cloud computer with its own browser, which runs independently of your local machine. Optional local-computer access exists as a separate, user-controlled capability.