ChatGPT Dots: What They Are, What They Cost, Limits
Dots are OpenAI’s always-on agents inside ChatGPT. You give your dot a goal, and it works toward it around the clock on its own cloud computer, using the apps you connect, then brings the result back for your review. OpenAI launched them at DevDay on September 29, 2026, on GPT-6 Astra, for Pro and Business Premium subscribers aged 18 and over.
Here is what a dot does, who can get one, what it costs, how to set it up, and the limits to plan around. If you are already weighing options, we ranked the best ChatGPT dots alternatives.
How a dot works
A normal ChatGPT chat answers and stops. A dot takes on an ongoing responsibility and keeps making progress between conversations. The mechanics OpenAI documents:
- It has its own computer. Each dot runs on a cloud computer with its own browser, separate from your machine, and you can open it any time to inspect the work.
- It runs several projects at once, so you keep handing it tasks without juggling threads.
- It reaches over 4,000 apps through ChatGPT plugins, using the connections and permissions you already set in ChatGPT.
- It can use your laptop, if you let it. Local access starts off. Once allowed, the dot can work with your files and fall back to your own browser when its cloud browser is blocked.
- It hands work to Codex and ChatGPT Work, creating tasks there on your behalf.
- It learns how you work. It starts with your ChatGPT memory, forms its own memories from connected apps, and adjusts to your feedback.
- It looks for work while you are away. OpenAI calls this proactive research: background tasks that read your connected apps with read-only tools and cannot send messages, change app content or control a computer.
You message or voice-call your dot in ChatGPT on desktop, web and mobile, and message it in Slack and Microsoft Teams, with context carried across all of them. OpenAI’s lead example: an early tester’s dot noticed he had forgotten to invoice a publication, prepared the invoice, and sent it once he approved.
What a dot is allowed to do on its own
Before a dot sends an email or changes a file, a separate safety system called Auto-review checks the planned step against your instructions, your rules and OpenAI’s safety requirements. For an email, it checks the recipient and the content for a wrong address or something you did not mean to share.
Some rules are fixed. Purchases use cards already saved on the merchant’s site and always need your approval. Permanently deleting data or granting new security-sensitive access needs your confirmation every time. Changing a password or moving money between financial accounts is handed back to you. On supported sites, saved passwords are submitted to the dot’s browser without the model seeing them.
Everything else you shape with Custom Rules: describe an action, then pick Take action without asking, Take action if pre-approved, Ask before taking action, or Hand off to you. “Pre-approved” means you asked for that action explicitly in your prompt.
Who gets dots, and what they cost
| Plan | Monthly price | Dot included |
|---|---|---|
| Free | $0 | No |
| Go | $8 | No |
| Plus | $20 | No |
| Pro 100 | $100 | Yes |
| Pro 200 | $200 | Yes |
| Pro 500 | $500 | Yes |
| Business Premium | Per seat | Yes |
| Enterprise, Edu, Healthcare | Contact sales | Beta, off until an admin enables it |
Your first dot is included at no extra cost, and it is the only one for now. OpenAI says you will later be able to add more and scale each one’s speed or monthly workload.
Talking to your dot does not count toward your ChatGPT limits; tasks it starts in Codex or ChatGPT Work do. For the first month, dots usage will not count toward plan allowances at all, and OpenAI will publish usage terms for each plan after that.
The Pro tiers changed the same day. Pro 500 is new, with 25 times the Plus allowance and the only Pro access to Astra Ultrafast. Pro 200 reopened at a lower allowance, and OpenAI emailed existing subscribers that from October 30, 2026 their included ChatGPT Work and Codex usage drops from 20 to 10 times Plus, at the same $200. Details are in our September 29 evening AI news roundup.
Region matters as much as plan. Pro subscribers in the European Economic Area, Switzerland and the UK do not get dots at launch. Business Premium gets them in every region ChatGPT supports. Rollout is gradual and can take several days to reach an account.
How to set up your dot
- Open ChatGPT on a computer. Use the ChatGPT desktop app (Mac or Windows) or chatgpt.com in a desktop browser. You cannot create a dot on mobile, and mobile web is not supported.
- Follow the onboarding prompts and name it. The default handle is @yourname-dot; naming it changes that to @yourname-agentname. Pick a character or a pet as its avatar.
- Connect your apps. Use the Plugins section of ChatGPT Settings (on mobile, the dot’s profile under Customize → Plugins), choose an app and review its permissions before connecting. These are your shared ChatGPT connections.
- Set your Custom Rules. In the ChatGPT mobile app, Customize → Custom rules lets you review the defaults and add rules for anything you want blocked, pre-approved or always handed back.
- Add channels and your computer. Slack (and texting, if you have the beta) is set up from desktop. To let the dot work locally, open the desktop app on that machine and confirm Allow access.
- Give it a task. Describe the job and the details it needs. For recurring work, ask directly: “review my calendar each morning and tell me what’s coming up.”
After setup you can talk to it from the ChatGPT mobile app. Its profile shows work under In progress, Scheduled and Completed, and Pause in the ••• menu halts it until you resume.
The limits to plan around
The floor is $100 a month. There is no Plus path. Free, Go and Plus subscribers do not get a dot, so the cheapest way in is Pro 100 or a Business Premium seat.
Personal plans in Europe and the UK are out. Pro is excluded in the EEA, Switzerland and the UK at launch, with no date given.
It has no email address of its own. You connect your personal email for the dot to use. At launch you cannot give it its own standalone address, so nobody can cc it on a thread and everything it sends goes out as you.
It cannot call you, and texting is barely there. Texting is a limited beta for some US Pro users, through a third-party provider, and not available in Business or Enterprise workspaces.
It starts on your schedule, not on events. OpenAI documents three ways a dot starts work without a new message: reminders, recurring checks, and read-only background research. It does not describe a dot waking up because a form was submitted or a deal changed stage. Separately, ChatGPT automations gained event triggers from supported plugins today; OpenAI has not said whether a dot can own one.
Forgetting is all or nothing. Disconnecting an app does not delete what the dot already learned from it. The only way to delete its memories is Reset, which deletes the dot, its conversations and its scheduled tasks.
It is day one. OpenAI’s own guidance is that dots can make mistakes, including when following your rules, so review consequential work.
Dots vs Meta Muse
Meta Muse launched three weeks earlier on the same design: an agent on its own cloud machine that stops for approval on sensitive actions. Muse starts free, tops out at $100 and is limited to the US and Canada. A dot starts at $100, lives in ChatGPT, Slack and Teams, and leans toward work run through Codex and ChatGPT Work.
When a dot is the right call
A dot fits if you already pay for Pro, your work runs through ChatGPT, Codex, Slack or Teams, and you want a project kept moving between check-ins: a deck kept current, a migration tracked, feedback turned into fixes.
It is the wrong fit if you are on Plus, on a personal plan in Europe or the UK, or need an agent clients can email directly that starts on its own when something happens in your business. For those, see the ChatGPT dots alternatives.
The same idea, with its own inbox
The premise is right: an agent that owns an outcome beats a chat window you keep prompting. Carly builds that around the places work arrives.
- Every agent gets its own email address. Build AI employees for intake, scheduling or billing, and clients and colleagues cc them like a person. On a business account each one can also get its own phone number.
- It starts on its own. An email comes in, a meeting gets booked, a lead changes stage in your CRM, a row lands in a spreadsheet, and the workflow runs within minutes without anyone asking.
- You set how far it goes. Each agent sends on its own or drafts for your approval, and reaches only the tools you give it.
- It reaches the systems the work lives in. Carly connects to thousands of apps.
- It meets you where you are. Use it on the web or text Carly from your phone. If you live in ChatGPT, add the Carly plugin there too.
- It starts well below $100. Free Zapier-style workflows; AI agents from $35/month.
If you want an agent with its own inbox working this week, get started with Carly.
FAQ
What are ChatGPT dots? Always-on AI agents in ChatGPT, launched by OpenAI on September 29, 2026. Each dot runs on GPT-6 Astra with its own cloud computer, connects to your apps through ChatGPT plugins, and keeps working toward your goals between conversations.
How much do ChatGPT dots cost? Your first dot is included with ChatGPT Pro ($100, $200 or $500 a month) or a Business Premium seat. For the first month dots usage does not count toward plan allowances; OpenAI will publish usage terms after that.
Can I get a dot on ChatGPT Plus? No. Dots are not on Free, Go or Plus. The cheapest route is Pro 100 at $100 a month.
Are dots available in the UK and Europe? Not on Pro, which excludes the European Economic Area, Switzerland and the UK at launch. Business Premium users get dots in all supported ChatGPT regions.
How do I create my dot? On a computer, in the ChatGPT desktop app or a desktop browser. Name it, connect your apps, and set your Custom Rules. You cannot create one on mobile, but you can talk to it there afterwards.
Can my dot send email? It can use your personal email account once connected, with sends checked by Auto-review. At launch a dot cannot have its own email address, so messages go out as you.
Why are they called dots? OpenAI has not explained the name. Before launch, references to the agent turned up in ChatGPT’s configuration under the name “o”.
More: ChatGPT dots alternatives · Meta Muse · ChatGPT plugins · ChatGPT scheduled tasks · Best AI employees · Best AI personal assistants
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
