ChatGPT Dots: The Short Version
ChatGPT just changed again — and this time it’s a big one. OpenAI has launched Dots: a proactive, always-on AI assistant living inside ChatGPT. Not a chatbot that waits for your questions, but an agent that works in the background with its own cloud computer, browser, memory, and plugins.
If that sounds like science fiction, you’re not alone. Let’s break down what Dots actually are, what they can do, and whether you should care.
Announced at OpenAI’s DevDay 2026 (late September 2026), Dots are OpenAI’s new agentic assistants built into ChatGPT. Think of each Dot as a little AI coworker: it can browse the web, remember context about you, use plugins and apps, and — here’s the new part — act on its own instead of waiting for you to prompt it.
The regular ChatGPT you know is reactive: you ask, it answers. Dots are meant to be proactive: handling tasks, checking things, and getting work done while you do something else.
How Are Dots Different From Regular ChatGPT?
This is the key shift, so let me make it concrete:
- Regular ChatGPT: you type a question, it replies, conversation over. Every task starts with you.
- Dots: you give it an ongoing job — say, “keep an eye on flight prices for my December trip” or “summarize my team’s updates every morning” — and it works on it continuously, checking back with you when there’s something worth knowing.
Under the hood, OpenAI gives each Dot its own cloud computer, a browser it can operate, long-term memory, and access to plugins and apps. That’s the infrastructure that lets it do multi-step work instead of just chatting.
What Can Dots Actually Do?
Based on OpenAI’s announcement and early coverage, Dots are designed to:
- Handle multi-step tasks — research something across several sites, compare options, and report back with a summary
- Work proactively — monitor things over time and notify you, instead of you having to ask repeatedly
- Remember context — build up memory of your preferences and past work so you don’t repeat yourself
- Use apps and plugins — this ties into OpenAI’s bigger push: ChatGPT is becoming a platform where you find and use apps, not just a chatbot
- Delegate — hand a Dot several tasks at once and let it work through them
Early testers describe it as feeling like a “chief of staff” — you delegate, it executes, you review. That’s the pitch, anyway.
How to Try ChatGPT Dots
Dots are rolling out as part of ChatGPT’s latest updates. Here’s the honest guidance:
- You need a paid plan — Dots are only available on ChatGPT Pro or Business Premium, not on Free, Go, or Plus
- Update your ChatGPT app (or open chat.openai.com) and look for Dots in the interface
- If you don’t see it yet, it’s likely still rolling out — OpenAI typically releases big features in waves, not to everyone on day one
- Start small: give a Dot one simple ongoing task before trusting it with anything important
Good news for Pakistan: OpenAI’s initial exclusion list covers only the EEA, Switzerland, and the UK — Pakistan isn’t excluded, so a Pro subscription from a Pakistani account should get Dots once the rollout reaches you.
I won’t pretend there’s a magic trick to unlock it early — if it’s not in your app yet, waiting for the rollout is the real answer.
Dots vs Meta Muse vs Grok Bot
You’ve probably noticed every AI lab is racing toward the same idea: Meta has Muse, xAI has its own agents, and now OpenAI has Dots. The Verge even framed Dots as OpenAI’s Muse competitor.
Honestly? It’s too early to crown a winner. Each one lives inside its parent ecosystem — Dots inside ChatGPT, Muse inside Meta’s apps — so the “best” one will mostly depend on where you already spend your time. What matters for you: the era of the passive chatbot is ending. Every major assistant is becoming an agent that does things, not just says things.
The Honest Caveats
Let me balance the hype, because Dots are brand new:
- It’s version one. Early coverage mentions frustrations alongside the wins — expect rough edges, missteps, and tasks it fumbles.
- Memory cuts both ways. A Dot that remembers everything about you is convenient — and also a privacy consideration. Check what’s stored and manage it deliberately.
- Proactive isn’t always welcome. An assistant that acts on its own needs good boundaries. Start with low-stakes tasks until you trust its judgment.
- Availability varies. Rollouts are gradual; not everyone gets new ChatGPT features at the same time.
Frequently Asked Questions
What are ChatGPT Dots?
Proactive AI agents inside ChatGPT with their own cloud computer, browser, memory, and plugins — designed to work on tasks continuously, not just answer questions.
Are ChatGPT Dots free?
No — Dots need a paid plan. OpenAI has confirmed Dots are only on ChatGPT Pro ($100, $200, and the new $500/month tiers) and Business Premium ($125/user/month). Free, Go, and Plus plans don’t get Dots. Your first Dot is included at no extra charge, and chatting with it doesn’t eat into your usage allowance — though Work and Codex tasks it runs do count toward it. Pricing for extra Dots beyond the first hasn’t been announced yet.
How are Dots different from GPTs (custom chatbots)?
Custom GPTs are specialized chatbots you talk to. Dots are agents that act — they work in the background, use tools, and come back with results.
When will I get Dots?
It’s rolling out in waves. Keep your app updated and check periodically.
Should I trust a Dot with important work?
Not yet. Treat it like a talented new intern: great for drafts and research, but review everything before it matters.
The Bottom Line
Dots are OpenAI’s biggest signal yet that ChatGPT is growing from a chatbot into a work platform — one where AI agents handle ongoing tasks while you get on with your day. It’s early, it’s rough around the edges, but the direction is clear.
If you’re a ChatGPT user, this is worth watching closely. And if Dots show up in your app — try one small task and see for yourself.