Skip to content

News

OpenAI Introduces ‘Dots’ AI Agents for Proactive Research

OpenAI's new 'Dots' agents bring 'always on' proactive and scheduled read-only research, enhancing monitoring for Pro and Business Premium users.

OpenAI's green dot Todd beside a ChatGPT conversation, analysis chart, and email draft.
Image: OpenAI. Image: Search Engine Journal

OpenAI launched new artificial intelligence agents called ‘Dots’ on Tuesday, offering proactive, read-only research capabilities to autonomously monitor connected data sources.

The new agents are designed to continuously check for updates without direct user prompts, expanding the functionality of existing ChatGPT scheduled tasks by maintaining context across multiple conversations, according to the company.

Dots can perform recurring checks on a predefined schedule, similar to the scheduled tasks available in ChatGPT, OpenAI said. This feature allows the agents to monitor information streams for changes or new data over time.

The proactive research conducted by Dots is strictly read-only, meaning the agents can access and interpret information and save private notes, but they are unable to send messages, modify content via plugins, or directly control a web browser or computer.

OpenAI is rolling out Dots to Pro users in various markets, excluding the European Economic Area, Switzerland, and the United Kingdom. Business Premium users across all supported ChatGPT regions will also gain access to the new feature.

Enterprise, Education, and Healthcare workspaces can access a beta version of Dots, which is disabled by default. These organizational users will have administrative controls to enable or disable the feature, OpenAI said.

Dots leverage existing plugin connections and establish their own memory for storing information. Disconnecting an application will stop new access but will not erase previously saved memories, the company stated.

Users currently lack the ability to view or edit individual Dot memories. However, resetting a Dot will delete all associated conversations, saved memories, and scheduled tasks.

During assigned tasks, a Dot can operate its own browser and cloud computer. Enterprise administrators retain the ability to enable or disable these specific functionalities for their workspaces.

OpenAI indicated that future updates will enable users to add more Dots and scale their output. The Enterprise release of Dots is currently in a beta phase.

The company highlighted the potential for Dots in various monitoring responsibilities, particularly for detecting changes in search marketing elements or other dynamic online content.


Need links for your own site?

Guest posts from $59, link insertions from $45, or a managed plan from $269/mo.