OpenAI launches dots, always-on ChatGPT agents with their own cloud computer

OpenAI launched dots on 29 September 2026: always-on agents in ChatGPT, powered by GPT-6 Astra, that each run on their own cloud computer and keep working toward a user's goals 24/7. Dots are rolling out first to ChatGPT Pro and Business Premium users in eligible markets, with an admin-enabled beta for Enterprise, Edu and Healthcare workspaces. For marketing teams, dots are built to take on ongoing work, such as content repurposing or launch revisions, in the background rather than one prompt at a time.

OpenAI set out the details in its announcement, Introducing dots, published the same day as its DevDay 2026 recap. A companion OpenAI Help Center article, "Getting started with your dot", adds the plan and market rules. This article covers what dots do, the controls OpenAI built around them, who can use them now, and what the announcement leaves open.

What OpenAI announced on 29 September 2026

OpenAI describes dots as always-on agents. Each dot is powered by GPT-6 Astra, has its own cloud computer and browser, learns from feedback over time, and can work toward a user's goals around the clock. Through OpenAI's plugin ecosystem, OpenAI says a dot can connect to over 4,000 apps.

The difference from a normal ChatGPT chat is continuity. According to OpenAI's Help Center, a dot can take on ongoing responsibility and keep making progress between conversations. The user gives it a goal, defines what it may do on its own, and the dot brings results back for review, turning to the user for decisions that need human judgment.

OpenAI also says a dot can take a project and run with it while working on several others, so users do not have to juggle separate threads. Users can open the dot's cloud computer at any time to inspect its work. Access to the user's own laptop is optional: the Help Center says it starts turned off and must be allowed from the ChatGPT desktop app.

Where you talk to a dot

OpenAI says users can message or call their dot in ChatGPT on desktop, web and mobile, and message it in Slack and Microsoft Teams. Texting is described as coming soon. The Help Center narrows that: texting is a limited beta for Pro users in the US only, and it is not available in Business or Enterprise workspaces. The Help Center also says a dot must be created in the ChatGPT desktop app or on desktop web, not on mobile, and that a dot cannot initiate calls to the user at launch.

Dots at a glance

The table below summarizes the main facts from OpenAI's 29 September 2026 announcement and its Help Center article.

How OpenAI says dots stay under control

Most of OpenAI's announcement is about limits rather than features, which matters for any business handing an agent access to its accounts. OpenAI lists several layers.

Proactive research is read-only

When a user is not actively working with it, a dot looks for ways to help in the background. OpenAI calls this "proactive research". It uses the apps the user has already connected, but through tools restricted to read-only, which OpenAI says cannot send messages, change app content or control the browser or computer. The Help Center adds that a dot can form memories from connected apps during this background review, and that disconnecting an app does not delete what the dot already learned from it; deleting that information requires resetting the dot.

Custom Rules and auto-review

Dots start with built-in rules for when to act alone and when to ask. Custom Rules let users allow specific actions, require approval for them, or block them. In the Help Center version, the choices are to take action without asking, take action if pre-approved, ask before taking action, or hand off to the user. OpenAI says dots also use auto-review to check any action that could affect a user's accounts or share information against the user's instructions, Custom Rules and safety requirements. Certain sensitive tasks, such as changing a password, always stay with the user.

Monitoring and data use

OpenAI says a monitoring system can pause or stop a dot's work if it detects a safety concern, and that progress, including background work, is visible in Activity View. On training, OpenAI says it does not use content from ChatGPT Business, Enterprise or Edu workspaces to improve its models by default, that users on personal plans can control whether a dot's work is used, and that it does not train directly on proactive research or a dot's notes to itself. OpenAI also says plainly that dots can still make mistakes and that consequential work should be reviewed.

The marketing use cases OpenAI described

OpenAI gave five example roles for a dot. Two are directly about marketing work.

• Content production. For a content creator, OpenAI says a dot learns the creator's voice and audience. When a new interview transcript arrives, it identifies moments for clips, prepares show notes and drafts social posts for the creator's approval, then carries the creator's edits across the materials.
• Launch revisions. For someone leading a product launch, OpenAI says a dot learns the audience, positioning and creative standards. When the product scope changes, it works out what that means for the story, revises launch materials and prepares drafts of asset and document changes for review.

The other examples cover a developer turning customer feedback into tested fixes, a scientist rerunning analyses as new data arrives, and a sales lead updating an enterprise proposal as requirements shift. OpenAI also mentions an early tester whose dot noticed a forgotten invoice, prepared it and sent it after his approval.

In every marketing example OpenAI gave, the dot drafts and the human approves. Nothing in the announcement describes a dot publishing content or spending ad budget without a person signing off. That reading comes from the examples OpenAI chose, and teams should still set their own Custom Rules rather than assume the default fits.

Specialist dots and Microsoft Agent 365

Alongside personal dots, OpenAI is previewing specialist dots. These take on a fixed responsibility inside an organization, with their own identity, credentials and access to the systems they need. OpenAI says it is building on early internal testing across procurement, invoice processing, email marketing, customer support and commercial contracting.

Specialist dots are starting as focused enterprise pilots, with OpenAI's engineering teams working directly with each organization to define the dot's responsibilities, tools and review process. OpenAI is also working with Microsoft to integrate specialist dots with the governance and security controls in Microsoft Agent 365, so that businesses can manage dots through Microsoft tools they already use. No date was given for that integration.

Plans, pricing and usage limits

OpenAI says a user's first dot is included in the Pro or Business Premium plan at no extra cost, with an allowance for deeper work and extended limits for the first month after launch. Conversations with a dot do not count toward ChatGPT usage limits. Tasks the dot starts or manages in Codex or ChatGPT Work do count toward limits as usual.

OpenAI says that in the future users will be able to add more dots and scale each dot's output by increasing its speed or the amount of work it takes on per month. The announcement did not give a price for extra dots or for that added capacity. Some third-party coverage has reported figures, but they do not appear on OpenAI's announcement or Help Center page, so they are not repeated here.

What OpenAI did not say

• A named country list. OpenAI's Help Center names only the exclusions for Pro (the European Economic Area, Switzerland and the UK) and says Business Premium covers all supported ChatGPT regions. It does not publish a country-by-country list, and it does not mention Thailand by name.
• Exact rollout timing. The Help Center says dots are rolling out gradually and access may take several days to reach an account.
• Prices beyond the first dot. Not stated.
• Which of the 4,000-plus apps matter for marketers. OpenAI did not list them. Users see available apps under the dot's plugin settings, and the Help Center notes that available apps may differ.
• Messaging apps used in Asia. The channels OpenAI named are ChatGPT, Slack, Teams and, in a US-only beta, texting. LINE is not mentioned.

What this means for Thai marketers

On availability, OpenAI's own wording is the safest guide. Pro is offered outside the EEA, Switzerland and the UK, and Business Premium covers all supported ChatGPT regions. Thailand is not among the named exclusions, but OpenAI does not list it either, and rollout is gradual. A Thai team on Pro or Business Premium should check its own account rather than rely on reports.

For agencies and in-house teams, the practical question is governance. A dot with read access to email, drive, Slack and analytics can form memories from that data, and disconnecting an app does not erase them; only resetting (deleting) the dot does. Before connecting client or company accounts, it makes sense to decide which apps a dot may read, which actions need approval, and who reviews drafts. Thai teams that coordinate mostly in LINE will also find that the channels OpenAI supports today are ChatGPT, Slack and Teams.

There is a search angle too, framed here as analysis. Dots browse the web with their own browser and research in the background on a user's behalf. As more research is done by agents rather than people clicking through results, a brand's content has to be easy for AI systems to read, trust and cite. That is the work of generative engine optimization and of making sure your brand appears when people ask ChatGPT for recommendations, often called ChatGPT SEO.

Frequently asked questions

What are OpenAI dots?

Dots are always-on agents in ChatGPT, launched by OpenAI on 29 September 2026, that run on GPT-6 Astra with their own cloud computer and browser. They can connect to over 4,000 apps through plugins, keep working between conversations, and bring results back for the user's review.

Are dots available in Thailand?

OpenAI has not named Thailand specifically. Its Help Center says Pro access excludes only the European Economic Area, Switzerland and the UK, and Business Premium covers all supported ChatGPT regions, with a gradual rollout that may take several days to reach an account.

How much do dots cost?

The first dot is included in ChatGPT Pro and Business Premium plans at no extra cost, according to OpenAI. OpenAI has not published prices for additional dots or extra capacity.

Can a dot send emails or post on my behalf without asking?

It depends on the rules you set: Custom Rules let you allow an action, require approval, or block it, and auto-review checks actions that could affect your accounts. Background proactive research uses read-only tools that OpenAI says cannot send messages or change app content.

Do conversations with a dot use up my ChatGPT limits?

No, OpenAI says conversations with a dot do not count toward ChatGPT usage limits. Tasks the dot starts or manages in Codex or ChatGPT Work do count as usual.

The bottom line

Dots move ChatGPT from answering prompts to holding a standing assignment, and OpenAI has put most of its launch effort into the rules around that: read-only background research, Custom Rules, auto-review and a monitoring system that can stop a dot. For marketing teams, the sensible first step is a small, approval-gated trial on low-risk work such as content repurposing, with clear limits on which accounts a dot can read. If you want to understand how agents like dots will find and judge your brand online, Relevant Audience's GEO team can review how visible your content is to AI systems.