Start Here

Start talking. Work it out together.

Dot is OpenAI’s ongoing personal assistant in ChatGPT. You can begin with an unfinished thought, find a useful next step, and return when your situation changes.

A light first step.

  1. Say what’s on your mind. A goal, a loose end, or “I don’t know where to begin” is enough to start.
  2. Find the next step together. Answer a question, share a relevant note, and ask what would be useful next.
  3. Check one result and make a correction. Compare it with what you know. Add a missing detail or correct an assumption. Come back when the priority changes.

You don’t need a perfect prompt

Find a first message.

Choose a starting point. Add what’s on your mind.

My idea is still a little scattered

I have a goal in mind, but my thoughts are scattered. Help me work out a useful next step. Ask me one question at a time.

I want to keep a project moving

Help me keep this project moving. I’ll share my notes and where things stand. Help me choose the next step and suggest a useful follow-up as things change.

I’m not sure where you could help

Here are a few things taking up my attention. Help me choose where you could be useful, and ask what information or access you need and why.

Copy a message into your conversation with dot and make it your own. You can work out the details together.

Need to set up your dot first?

Use the latest ChatGPT mobile app, the desktop app, or a desktop browser. Open dots and follow the introduction; you can name it and change its appearance. Mobile web is not supported. Plugins and a connection to your personal computer are optional—you can start talking first.

Access depends on plan, rollout, region, and workspace policy. An eligible account may still be waiting. See the official setup guide and current availability. Reviewed October 14, 2026.

You can steer in everyday language.

A useful conversation leaves room for you to change your mind. Try one of these when you return:

  • “That’s too much detail”
  • “My priority has changed”
  • “What’s finished and what’s blocked?”
  • “What information or access do you need?”

Something stuck? When your dot needs steering gives you a useful thing to say, a check to make, and what to expect next.

Dot uses available conversation context and saved notes to carry work forward. That isn’t perfect memory: remind it of an important fact, name the source, and correct anything it has misunderstood.

Based on OpenAI’s tasks and memory guide.

When work involves access, actions, or important results

Ask what information or access would help and why. Available connections and controls depend on your account. A conversational instruction or starter brief doesn’t connect an app, change a setting, or override a safeguard.

Start with preferences you’re comfortable with, such as “show me drafts before sending.” You can explicitly authorize a defined action or ongoing responsibility later. Check the scope and the actual product permissions together.

For important actions, ask dot to show the result or confirmation, and check it before relying on it.

Read our connections, boundaries, and feedback lessons, or OpenAI’s controls guide and privacy FAQ.

A helper you can keep talking to.

Imagine a helper with a notebook. You bring a half-formed plan. Together you work out what matters, what would help, and what to do next. Later, you return with an update.

That’s one way to picture dot. The notebook represents selected context and saved notes, not a perfect record of everything you’ve said. It can miss details, so bring important changes back into the conversation.

Our analogy, based on OpenAI’s getting-started guide.

The AI words, in everyday language.

These words describe different parts of the picture. One product can be a chatbot, use an LLM, and run agents and automations.

LLM / model

The language engine. A large language model learns patterns from training data to work with language and generate responses. It can make mistakes.

An ingredient inside an AI product, rather than the whole product.

Chatbot

A way to talk with AI. You ask, it replies, and you steer the conversation. Some chat products also have memory, tools, and scheduled tasks.

“Help me think of a good menu for our neighborhood picnic.”

Agent

AI that can work toward a goal across several steps, using tools when available. How much it can do depends on the product and its permissions.

“Compare three picnic locations and prepare a shortlist.”

Automation

Work set up to run when something happens or at a chosen time. A simple automation follows fixed steps; more flexible ones can include an AI agent.

“Every Friday, remind me to check the guest list.”

Tools / connections

Ways an AI product can read information or do something beyond writing a reply. Each connection has its own access and permissions.

Access to a planning document, an app, or a browser.

Dot

OpenAI’s ongoing personal assistant in ChatGPT: an agent you can talk with, give work to, and return to as things change.

“Help keep the picnic plan up to date as we get ready.”

What makes Dots different?

A Dot is a kind of agent. Its emphasis is an ongoing relationship with your work. OpenAI describes a helper with saved context, its own cloud computer, connected tools, and the ability to continue between conversations.

You can give it new information or change direction as the work develops. It can follow up without every next step needing a fixed timer. It can also delegate work to other agents. See how tasks and memory work.

Other AI tools can also remember things, use tools, and do recurring work. The appeal of Dots is how those pieces come together in one continuing helper. It isn’t a promise that Dots is better at every task.

Why might I want one?

Because remembering to follow up can be a job of its own. You might want help gathering updates, keeping a plan current, or preparing the next draft while you focus on the people involved.

A useful fit: something you want to keep moving as the information or your priorities change.

A simpler tool may be enough: use a chat for a one-off explanation, a reminder for a deadline, or fixed automation for a predictable sequence of steps.

A simple conversation can help you discover where it fits. The available context, connections, and permissions shape what it can do.

How ongoing help, schedules, and permissions fit together

You can give dot a defined responsibility and say which changes deserve an update. For a check-in at a fixed time, agree on the timing, time zone, and end date, and ask it to confirm the saved schedule. A recurring-help request written on this website does not create one.

Dot may also notice a useful connection in information it is allowed to read and suggest a next step. That background research is read-only; taking action afterward needs the relevant authorization and permissions. Read OpenAI’s guide to controlling your Dot.

Where a service supports event monitoring, you can ask dot to watch a specified event and confirm the setup. Connecting a source alone does not create a monitor. See the steering guide if a schedule or monitor is unclear.

The hope of Dots: what could be.

Two fictional stories showing possible ways to work together. These are illustrative possibilities, not customer testimonials, tested results, or guaranteed behavior. Timing and usefulness vary with context, access, permissions, and the situation.

Maya brings a messy plan back into focus.

Maya repairs bicycles. Her first message is a jumble: parts to order, customers to call, and a website she wants to improve. Dot asks what matters most today, then helps her sketch a plan from the shop notes she shares.

She returns later: “Two urgent repairs came in. Pause the website work.” Dot revises the plan. It has one pickup down for Thursday; Maya corrects it to Wednesday, and it changes the order of work.

Maya authorizes dot to keep a private repair checklist current using the selected job sheet. She also agrees to an afternoon check for missing parts that week, with a saved schedule and updates when something needs her attention.

Maya decides which customer commitments and purchases to authorize.

The hope: a plan that changes with the day, without rebuilding it from scratch each time.

Explore the follow-up starter

Jon gets to enjoy the picnic he’s organizing.

Jon volunteers to organize a neighborhood picnic. He starts with a rough idea, a budget, and an uncertain headcount. Dot helps turn his notes into a workable first plan.

When the venue changes, Jon comes back with a lower budget and a new headcount. Dot revises the supply list and highlights the decisions that still need him. They agree to a Tuesday check-in until the event and confirm the schedule.

Later, new venue information becomes available in a document dot is authorized to read. It could connect the rental fee to Jon’s budget limit and suggest revisiting the plan. That kind of helpful suggestion can arise from context, alongside the agreed schedule; its timing and usefulness aren’t guaranteed.

Jon checks the counts and chooses what to change. In this arrangement, he keeps purchases, reservations, and invitations for himself.

The hope: a plan that’s easier to pick up again, with fewer small details living only in Jon’s head.

Explore the event-planning starter

Product descriptions checked October 14, 2026 against the official guides linked above. The stories and analogies are our own. We have not tested these scenarios in Dots.

Can I use Dots yet?

Access is rolling out. Availability depends on your plan, region, and workspace settings. Check your account and the current official setup guide.

Check access and setup

What do we do?

Just the Dots is an independent guide by Blisster. We offer starting messages, lessons, and an optional brief builder to help you find your own way of working with dot.

About this independent guide

A framework when you want one.

Our Five Dots framework helps you think through context, connections, responsibilities, boundaries, and feedback. You can start talking before you have all five answers. Return to the lessons when a question becomes useful.

If you don’t have access yet, explore the examples and save a starting message or brief. This site needs no OpenAI account. Our six playbooks are illustrations, not installed agents or hands-on tested workflows.