{"schemaVersion":"1.0","type":"Article","types":["Article"],"slug":"openai-s-dots-an-ai-agent-that-keeps-working-after-you-log-off-pi0kc","url":"https://zyvop.com/openai-s-dots-an-ai-agent-that-keeps-working-after-you-log-off-pi0kc","title":"OpenAI's Dots: An AI Agent That Keeps Working After You Log Off","subtitle":"OpenAI's new dots are always-on agents with their own cloud computer. Here's what they do, where you can reach them, and how the controls work.","tldr":"OpenAI launched dots on September 29, 2026: agents built on GPT-6 Astra that run on their own cloud computer, connect to over 4,000 apps, and keep working in the background.","keywords":["ChatGPT","OpenAI","gpt-6 astra","AI agents","dots","AI News"],"entities":["Sanju Singh","ChatGPT","OpenAI","gpt-6 astra","AI agents","dots","AI News","ZyVOP"],"keyTakeaways":["OpenAI launched dots on September 29.","Each one is an AI agent you can hand real projects to.","OpenAI calls it a new way to work with AI."],"headings":["What a dot is","What it looks like in practice","Where you talk to it","How the controls work","Specialist dots for companies","Who can get it, and what it costs","Before you try one"],"outboundLinks":["https://openai.com/index/how-we-build-safety-security-and-privacy-into-dots/","https://openai.com/index/introducing-dots/"],"contentText":"OpenAI launched dots on September 29. Each one is an AI agent you can hand real projects to. OpenAI calls it a new way to work with AI. You can still chat with a dot, ask it questions, or hop on a voice call, but you don't have to start every task yourself. Give it a name, connect your apps, and it keeps working on your projects in the background. What a dot is Dots run on GPT-6 Astra. Each dot has its own cloud computer and its own browser. It can use the apps you've connected, and OpenAI says its plugins link dots to over 4,000 apps. You can open your dot's computer at any time and check its work. You can also give it permission to use your laptop, so it can work next to you on your own machine. A dot can run several projects at once. You keep giving it new tasks without juggling separate threads. OpenAI says a dot learns from feedback over time. The longer you work together, the better it knows your preferences and what good work looks like to you. What it looks like in practice OpenAI shows five examples. In each one, the dot does the legwork and you review the result. Developer: The dot watches customer feedback for repeat requests, then builds and tests small fixes. You get complete PRs with videos of the changes. Product launch: When scope changes, the dot works out what it means for the story, revises launch materials, and drafts asset and doc changes. Scientist: As new data arrives, the dot reruns analyses, investigates odd results, and updates figures. It flags what needs your review. Sales lead: The dot checks customer requirements against product docs, builds a proof of concept for a key integration, and suggests a solutions engineer. It updates the proposal and test plan as things change. Content creator: From a new interview transcript, the dot picks clip moments, writes show notes, and drafts social posts for approval. OpenAI describes how its own teams use dots. A bug shows up in Slack and a dot starts investigating. A design arrives and a dot turns it into a working app. One early tester's dot noticed he'd forgotten to invoice a publication. It prepared the invoice and sent it after he approved it. Where you talk to it You can message or call your dot in ChatGPT on desktop, web, and mobile. You can also message it in Slack and Teams. Texting is coming soon. The dot carries the full picture across all of these, so you can start a project in ChatGPT, share context with your team in Slack, and let the dot follow through. It can message you with progress, questions, or decisions that need you. How the controls work Because a dot acts on its own, the controls matter. Here are the main ones: Separate computer: Your own computer and its contents stay separate from the dot unless you choose to connect it. Passwords: For supported sites, dots can sign in with saved passwords without exposing them to the model. Read-only background work: When you're not working with it, a dot looks for ways to help using your connected apps. OpenAI calls this proactive research. In this mode the tools are read-only, so the dot can't send messages, change app content, or control your browser or computer. Rules: Dots start with built-in rules for when to act alone and when to ask. Custom Rules let you allow, require approval for, or block specific actions. Activity View: You can follow progress, including background work, and redirect the dot. Auto-review: Actions that could affect your accounts or share information are checked against your instructions, your rules, and OpenAI's safety requirements. Some tasks, like changing a password, always stay with you. Monitoring: If OpenAI's monitoring system detects a safety concern, it can pause or stop the dot's work. On data, OpenAI says it doesn't use content from ChatGPT Business, Enterprise, or Edu workspaces to improve its models by default. On personal plans, you control whether your dot's conversations and work are used for that. OpenAI says it doesn't train directly on proactive research or a dot's notes to itself. It adds that information from them may be used if it helps inform an eligible conversation or task, depending on your settings. OpenAI says dots can still make mistakes, so review anything consequential. For more detail, OpenAI has a separate post on how it builds safety, security, and privacy into dots. Specialist dots for companies Your own dot works for you. A specialist dot takes on a defined responsibility inside a company. The company gives each specialist dot its own identity, credentials, and access to the systems it needs. OpenAI says it built on early testing inside its own teams across procurement, invoice processing, email marketing, customer support, and commercial contracting. This part is still early. OpenAI is starting with focused enterprise pilots, where its engineers work with each organization to define the dot's job, its tools, and how people review its work. OpenAI is working with Microsoft to connect specialist dots to Agent 365, so businesses can manage them with the governance and security tools they already use. Who can get it, and what it costs Dots began rolling out on September 29 in ChatGPT to Pro and Business Premium users in eligible markets. Enterprise users, including Edu and Healthcare, can try the beta once their workspace admin turns it on. Your first dot is included in your Pro or Business Premium plan at no extra cost. That plan also includes an allowance for deeper work, with extended limits for the first month after launch. Later, you'll be able to add more dots and scale each one by raising its speed or the amount of work it can take on each month. Conversations with your dot don't count toward your ChatGPT usage limits. If your dot starts or manages tasks in Codex or ChatGPT Work, those tasks do count. To start, create your dot in the ChatGPT desktop app or your desktop browser, connect your apps, and let it introduce itself. After setup, you can message it from the mobile app. Before you try one OpenAI says it wants teams of dots working together on your behalf over time. For now, each person starts with one dot, and OpenAI says you'll be able to add more later. Check whether your market is eligible, and decide which apps you're willing to connect. The rules and approvals are yours to set, so start narrow. Use Activity View to see what your dot is doing, then give it more room as you get comfortable. Source: OpenAI, \"Introducing dots,\" September 29, 2026.","contentHash":"sha256:0a5c2e4dcfacd363dc1ae66f6e6e2d91345585d0e9896f78bd8e6ce5aff1cf99","authorName":"Sanju Singh","authorUrl":"https://zyvop.com/author/sanjay687","authorSameAs":[],"category":"AI News","tags":["ChatGPT","OpenAI","gpt-6 astra","AI agents","dots"],"audience":"Readers and engineers researching AI News","tone":"Practical and evidence-based engineering guidance","readingTimeMinutes":5,"wordCount":1124,"faqs":null,"primaryTopic":"AI News","publishedAt":"2026-09-30T02:58:48.666Z","updatedAt":"2026-09-30T02:59:08.109Z","canonicalUrl":"https://zyvop.com/openai-s-dots-an-ai-agent-that-keeps-working-after-you-log-off-pi0kc"}