- OpenAI Dots release date: Dots are OpenAI's ongoing AI agents inside ChatGPT, launched with gradual rollout in 2026.
- Plan access: Pro plans get early access; Business Premium is broader; Enterprise requires admin enablement.
- Regional limits: At launch, Pro access excludes the UK, Switzerland, and the European Economic Area.
- Core capability: Each dot runs its own cloud computer, powered by GPT-6 Astra, and works between conversations.
- Best first step: Try one small responsibility with a clear finish line, then evaluate if it saved effort.
OpenAI Dots Release Date and Rollout Status
The OpenAI Dots release date marks the moment OpenAI moved ChatGPT from a conversation-based assistant toward an always-on agent model. Dots are ongoing AI agents inside ChatGPT, powered by GPT-6 Astra, and each one gets its own cloud computer. You assign a responsibility, connect the tools it needs, and set boundaries—the dot keeps the job moving between conversations.
OpenAI describes access as gradual rather than a single global launch day. If the feature button is missing in your account, waiting and checking back is the realistic path; refreshing aggressively does not change your tier.
Video Highlights:
- Explains what Dots are and how they differ from a normal chat
- Shows a launch-materials example: keeping drafts consistent with approved decisions
- Covers permissions, Custom Rules, and Auto-review action checks
- Details rollout limits by plan and region
| Rollout Factor | Status at Launch |
|---|---|
| Overall rollout | Gradual, not instant for all users |
| Pro plan | Included, but excludes UK, Switzerland, EEA |
| Business Premium | Broader regional availability |
| Enterprise | Requires an administrator to enable |
| Access method | Appears in ChatGPT once your account qualifies |
At launch, Pro access excludes the UK, Switzerland, and the European Economic Area. Business Premium offers broader coverage, and Enterprise users should ask their administrator to enable Dots before expecting the feature.
What Launches With Dots: Core Features
The release is not just a date—it introduces a working model for delegated, ongoing tasks. Each dot starts with ChatGPT memory and can build context from connected apps, which means it can connect new information to existing drafts instead of asking you to re-explain everything.
Ongoing Agents
- Own cloud computer per dot
- Continues work between chats
- Handles recurring responsibilities
Context and Memory
- Starts with ChatGPT memory
- Builds context from connected apps
- Respects your stated preferences
Background Work
- Researches in the background
- Runs scheduled checks
- Brings results back for review
| Feature | What It Does | Why It Matters |
|---|---|---|
| Custom Rules | Set boundaries like "draft only, never send" | Separates reading from acting |
| Auto-review | Action checks on sensitive operations | Some actions still need your approval |
| Scheduled checks | Dot re-checks sources before review | Catches contradictions automatically |
| Connected apps | Feeds context from your tools | Enables cross-document consistency |
Reading a document and emailing customers are different permissions. Access to the kitchen does not make someone the head chef—define what your dot may do before assigning responsibility.
Availability by Plan and Region
Understanding where the OpenAI Dots release date applies to you means checking three things: your subscription tier, your region, and whether an administrator controls your workspace.
| Plan | Access Level | Regional Notes |
|---|---|---|
| Pro | Early access | UK, Switzerland, EEA excluded at launch |
| Business Premium | Broad access | Widest coverage among business tiers |
| Enterprise | Admin-gated | Administrator must enable the feature |
| Free/Lower tiers | Not confirmed at launch | OpenAI says access is gradual |
OpenAI states access rolls out over time. A missing button does not mean your account is broken—it means your tier or region has not been reached yet. Watch official OpenAI channels for expansion updates.
Your First Dot: Step-by-Step Setup
Once the release reaches your account, start small. A good first dot has one clear responsibility, a defined source of truth, and a hard rule against sending anything.
Verify Access
Confirm Dots appears in your ChatGPT interface and that your plan and region are covered by the current rollout wave.
Define One Responsibility
Write a short job description, such as "keep these launch materials consistent with the latest approved decisions." Shorter than the meeting that created it.
Connect Context
Link the launch folder or relevant notes. Local computer access is optional and starts off—your entire laptop is probably not needed.
Set Rules and Boundaries
Add Custom Rules like "do not send anything" and configure Auto-review so sensitive actions return to you for approval.
Review and Verify
When the dot returns drafts, check the facts yourself. OpenAI's help page explicitly says dots can make mistakes, including when following rules.
First Dot Success Criteria:
- Keep three drafts consistent with approved notes
- Flag contradictions instead of guessing
- Bring back changes for review, never send directly
- Save more time than supervision costs
- Disconnect unused apps when the job ends
After the first run, ask whether the dot actually saved effort—including the time you spent supervising it. An assistant that needs another assistant is an expensive group project.
FAQ
Q: When is the OpenAI Dots release date for all users?
There is no single global launch date. OpenAI describes access as gradual: Pro users in eligible regions get early access, Business Premium has broader coverage, and Enterprise requires administrator enablement. Free-tier availability has not been confirmed at launch.
Q: Why don't I see Dots in my ChatGPT account?
The most likely reasons are plan tier, region, or rollout timing. At launch, Pro access excludes the UK, Switzerland, and the European Economic Area. If you are on Enterprise, ask your administrator to enable the feature.
Q: Can a dot send emails or messages on its own?
Only if you grant that permission. OpenAI provides Custom Rules and Auto-review action checks, and some actions still require your approval. A common starting rule is draft-only: the dot prepares changes and brings them back for review.
Q: Do dots make mistakes?
Yes. OpenAI's help page explicitly states dots can make mistakes, including when following rules. Always verify critical outputs yourself—the dot's confidence is not an acceptance criterion.
Q: What happens when I disconnect an app from a dot?
Disconnecting an app does not erase information the dot already obtained from it. Treat permissions thoughtfully rather than accepting everything at once, and disconnect tools when a job ends.