# OpenAI Dots Run Read-Only Research When Nobody Is Asking > Source: > Published: 2026-09-30 14:44:15+00:00 OpenAI’s Dots agents caught people’s interest at DevDay this week because of their “always on” feature. The help pages, however, didn’t get quite as much notice. They explain what a dot can and can’t do when it’s not being directly talked to, including checking the same sources for updates. OpenAI explains that a dot can [check its connected sources for new information even before anyone asks](https://help.openai.com/en/articles/20001529-dots-privacy-security-and-safety-faqs), which they call proactive research. It can also run regular checks on a schedule. The two methods follow different rules. Dots are now being [rolled out](https://help.openai.com/en/articles/20001530-getting-started-with-your-dot) to Pro users in various markets, excluding the European Economic Area, Switzerland, and the UK. They are also available to Business Premium users across all supported ChatGPT regions. Additionally, Enterprise, Edu, and Healthcare workspaces can try out a beta version, which is off by default. It might take several days for access to become available to your account. ## Proactive Research Runs Without A New Prompt The [launch post](https://openai.com/index/introducing-dots/) describes proactive research as what a dot does when the user isn’t actively working with it. The help center’s [privacy and safety page](https://help.openai.com/en/articles/20001529-dots-privacy-security-and-safety-faqs) spells out the limits: “Your dot can also look for useful updates before you make a request. This is called proactive research. It can read information from permitted connected sources and save private notes, but its research tools cannot directly: Send messages to other people. Change content through plugins. Control a browser or computer.” Every action the dot takes with what it finds follows the usual safety checks and rules. The Custom Rules, which are settings specific to each user for what a dot can do independently, allow for adjusting actions to be more or less flexible. However, they can’t override the proactive research restrictions, nor can the dot itself. The mode that operates without a prompt is limited to reading and taking notes. ## Recurring Checks Move Onto An Agent Scheduled work isn’t a new feature for ChatGPT. Its [help page on Scheduled tasks](https://help.openai.com/en/articles/10291617) already explains how ChatGPT can handle one-time or recurring tasks, keep an eye on changes, and respond to supported events from Gmail, Slack, or GitHub. A dot can perform those checks across apps connected by a user and maintain context between conversations. The example on the getting-started page shows a dot reviewing a calendar every morning and reporting upcoming events. Both scheduled runs and proactive updates happen within the same conversation. ## What The Dot Keeps Between Conversations A dot makes use of the plugin connections that a user has already set up. According to the [privacy and safety page](https://help.openai.com/en/articles/20001529-dots-privacy-security-and-safety-faqs), plugin permissions are shared across dots, ChatGPT, ChatGPT Work, and Codex. The [launch post](https://openai.com/index/introducing-dots/) also highlights that the plugin catalog now offers over 4,000 apps. The dot also creates its own memory from these connections. When you disconnect an app, it stops new access, but it doesn’t erase what the dot has already saved. Currently, users can’t view or edit these individual memories. Resetting the dot will delete its conversations, saved memories, and scheduled tasks. As of Sept. 30, none of the three Dots help pages mention an analytics or search platform. The examples of connected apps include a calendar, Slack, and a personal email account, and the [getting-started page](https://help.openai.com/en/articles/20001530-getting-started-with-your-dot) notes that available apps may vary. ## Dots Can Use Websites During Assigned Work While working on an assignment, a dot can operate its own browser and cloud computer. [Enterprise admins](https://help.openai.com/en/articles/20001554-manage-dots-in-chatgpt-workspaces) have the ability to enable or disable cloud browser use, cloud network access, and cloud computer use individually. Dots now sit alongside the [August WebMCP deployments](https://www.searchenginejournal.com/webmcp-connects-ai-agents-actions-inside-websites/587303/) from OpenAI, Shopify, and Cloudflare, which provided agents with structured methods to operate within websites. Neither the announcement nor the three support pages link Dots to search rankings or request website modifications. Instead, Dots introduce an additional agent capable of browsing a site on a user’s behalf. ## Why This Matters A dot can be assigned monitoring responsibilities, watch specific sources it is permitted to access, and report any changes. In search marketing, rankings, competitor pages, and campaign settings are checked periodically, with each check often starting anew. Proactive research is read-only, so a dot can detect a change overnight. Any follow-up actions are carried out following OpenAI’s action rules and safety checks. Whether a dot fits into a team’s monitoring system depends on which platforms it has permission to read through ChatGPT’s plugins. ## Looking Ahead OpenAI says users will later be able to add more dots and scale each one’s output, whether it’s speed or how much work it can take on each month. They also mention that the Enterprise release is in beta. The thing to watch is which analytics and SEO tools a dot can access through ChatGPT’s plugins, especially since proactive research only pulls from permitted connected sources. *Featured Image: OpenAI*