OpenAI launches Dots, an always-on AI assistant in ChatGPT

In short: OpenAI announced Dots, a new always-on AI assistant that can work across thousands of apps and report back inside ChatGPT.

What happened

OpenAI unveiled Dots during its DevDay keynote. The company describes Dots as assistants that stay active in the background and can handle tasks across connected apps, while learning what you prefer over time.

Each Dot uses its own “cloud computer,” which is like renting a computer on the internet that the Dot can operate for you. OpenAI says this lets Dots use a web browser and more than 4,000 supported apps. You can chat with a Dot in a text-message-like screen, or talk to it by voice through ChatGPT on the web, desktop, or mobile.

Dots can also connect with workplace tools like Microsoft Teams and Slack. OpenAI says they can carry over context, meaning your Dot can remember what you were doing on another device or app. For now, users can create one Dot, but OpenAI says it plans to support multiple Dots at once.

Why it matters

Tools like this are trying to turn AI from a chatbot you talk to into a helper that can actually do things for you. That can save time, but it also raises safety and privacy questions, especially after reports of similar assistants sharing sensitive information.

OpenAI says Dots include rules for when they can act on their own, and users can add custom rules, like forcing the Dot to ask permission before certain actions. OpenAI also says Dots have an “auto-review” check to confirm an action follows those rules, and to hand risky steps back to the user, like changing a password.

Dots start rolling out today to ChatGPT Pro and Business Premium users.

Source: The Verge AI