🚨 The "Always-On" Agent is Here: OpenAI Just Launched 'Dots' (And It Changes How We Code)

On September 29, 2026, OpenAI officially announced "Dots"—a new class of always-on AI agents that operate 24/7. Positioned as a direct rival to Meta's popular Muse agent, Dots are designed to handle ongoing projects, taking the concept of an AI assistant from a simple chatbot to an autonomous digital worker.

If you are a developer, this is a massive paradigm shift. You are no longer just sending a prompt and waiting for a single response; you are assigning a task to a background worker that has its own computing environment and continues to work while you step away.

Here is a breakdown of what Dots actually are, how the architecture works, and how they will impact your daily workflow.

🧠 What Exactly is a "Dot"?

Powered by OpenAI's new GPT-6 Astra model, a Dot is a named, persistent agent. Unlike traditional ChatGPT sessions that lose context or sit idle when you close the tab, Dots can continuously work toward your goals in the background.

Here are the standout features:

• Its Own Cloud Computer: Every Dot is provisioned with its own cloud computer and browser. You can open up your Dot's computer at any time to visually inspect its ongoing work.
• Massive Integrations: Through OpenAI's plugin ecosystem, Dots can connect to over 4,000 different applications.
• Multi-Channel Access: You don't have to keep a ChatGPT window open. You can interact with your Dot directly inside Slack and Microsoft Teams, or even hop on a voice call with it to talk through a problem. Texting capabilities are also coming soon.
• Continuous Learning: The more you interact and correct its work, the more your Dot learns your specific preferences, goals, and standards.

💻 How Developers Will Use Dots

Think about the tasks that currently drag down your velocity: tracking down a bug reported in Slack, updating documentation after a feature launch, or keeping an analysis current.

For example, when a bug appears in a Slack channel, your Dot can immediately start investigating the issue on its own. You could grant it access to your software project and feedback channels so it can trace the error, prepare a fix, and ask you for a review.

Conceptual API Integration

While Dots act as out-of-the-box personal agents, developers can leverage OpenAI's infrastructure, such as the Agents API which exposes a Codex harness, to programmatically trigger and manage background tasks.

Here is a conceptual example of how a developer might programmatically spawn a background task using a durable session API framework:

import { OpenAI } from "openai";

const client = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });

async function delegateBugFix(issueContext) {
  // Creating a durable background session for an always-on agent
  const session = await client.agents.sessions.create({
    agent_id: "dot-eng-primary", // Your dedicated engineering Dot
    goal: "Investigate the React rendering bug reported in #frontend-alerts",
    context: issueContext,
    tools: ["github_repo_access", "slack_read_write", "cloud_browser"],
    run_in_background: true
  });

  console.log(`Dot is investigating. Session ID: ${session.id}`);

  // The dot will message you on Slack or Teams when the PR is ready for review!
}

delegateBugFix({
  error_log: "Uncaught TypeError: Cannot read properties of undefined (reading 'map')",
  file: "UserProfile.tsx",
  timestamp: "2026-09-30T10:00:00Z"
});

Because the Dot has its own computing environment, it can theoretically clone the repo, run the tests, write the patch, and then just ping you for the final PR approval.

🔒 Security, Pricing, and Availability

Handing over the keys to your codebase and apps requires massive trust. OpenAI has implemented safety and privacy safeguards, allowing you to set strict boundaries, monitor progress, and remain the final decision-maker for critical actions. You can also give your Dot permission to securely connect to and use your local laptop directly alongside you.

Rollout Details:

• Dots are currently rolling out to users on Pro, Business Premium, and Enterprise plans in eligible markets.
• The first Dot is included at no extra cost on Pro and Business Premium subscriptions.
• OpenAI is also previewing "specialist dots" designed for enterprise systems, which come with IT-provisioned hardware, strict access management, and deep integrations into company systems of record.

🚀 The Bottom Line

We are officially moving from "AI as a Copilot" to "AI as a Colleague." By giving the AI its own persistent environment and the ability to work asynchronously, OpenAI is fundamentally changing what we consider to be a "developer tool."

What will you assign your Dot to do first? Let's discuss in the comments below! 👇