OpenAI Introduces ChatGPT Dots: 4000+ Apps Support, Availability, and More Explained

OpenAI has announced ChatGPT Dots, a new set of AI agents that can take a task off your hands and keep working without needing instructions at every step. You can give a Dot an initial instruction and leave it to work through the job, even after you leave ChatGPT. It can also work with connected apps in the background and come back when it needs your input or approval. OpenAI says Dots can take on everything from coding and research to sales, content creation and everyday tasks.

ChatGPT Dots Features

According to OpenAI, Dots run on GPT-6 Astra and each one gets its own cloud computer and browser. They can connect to more than 4,000 apps through OpenAI's plugin ecosystem, giving them access to the services needed for a particular task. After the user sets the goal, the agent can work through the next steps itself and handle several tasks at the same time.

OpenAI's examples include tracking recurring customer feedback, testing smaller code fixes, updating launch material, investigating new research data and revising sales proposals as requirements change. A Dot can also turn an interview transcript into clips, show notes and social posts, or handle everyday tasks such as comparing food delivery options and calculating the final bill before asking for approval to place the order.

A Dot can continue working in the background after the user leaves ChatGPT, returning with an update or asking for input when needed. The agent can be reached through ChatGPT, Slack and Microsoft Teams, while voice calls are supported and text messaging is coming later.

The cloud computer also keeps the agent separate from the user's own machine unless they choose to connect it. Dots can use another device when a task requires one, and users can watch what the agent is doing. On supported websites, saved passwords can also be used without exposing the credentials to the model.

Users decide how much access a Dot gets by connecting apps individually and choosing which actions it can handle on its own and which need approval. Background research stays read-only, so the agent cannot send messages or make changes to connected apps while working. OpenAI's auto-review system checks actions involving accounts or information sharing, while sensitive tasks such as changing a password still require the user to step in.

OpenAI says its safety systems can pause or stop a Dot if they detect malicious instructions or potentially harmful behaviour. The company also notes that Dots can make mistakes, so users should check their work before relying on it for anything important. Personal ChatGPT users can choose whether their Dot conversations and work are used to improve OpenAI's models, while Business, Enterprise and Edu content is excluded from training by default.

OpenAI is also testing Dots built for organisations, which can have their own identities and access selected company systems. Early uses include procurement, invoice processing, email marketing, customer support and commercial contracts. OpenAI is working with Microsoft to bring these agents to Agent 365.

ChatGPT Dots Availability

ChatGPT Dots are rolling out to Pro, Business Premium and Enterprise users in eligible markets. Enterprise users can access the beta when their workspace administrator enables it. OpenAI says the first Dot comes with Pro and Business Premium at no additional cost, with an allowance for deeper work and extended limits during the first month.

Conversations with a Dot do not count towards ChatGPT usage limits, although tasks that the agent starts or manages through Codex or ChatGPT Work count towards those services' limits. OpenAI says Dots are now available across ChatGPT on web, mobile and desktop, although the initial setup currently starts through the desktop app or desktop browser. More Dots and larger workloads are planned for the future.