MH-2795SEO & Search
OpenAI's Dots Can Research Before You Ask, But Only Read-Only
OpenAI's Dots agents can check connected sources for updates unprompted, but the proactive mode stays read-only — it cannot message, edit content or control a browser.
Wire notes
- Dots' proactive research is read-only: it can read permitted connected sources and save notes, but cannot send messages, change content through plugins, or control a browser or computer.
- Dots are rolling out to Pro users outside the EEA, Switzerland and the UK, to Business Premium users in all supported regions, and as an opt-in beta to Enterprise, Edu and Healthcare workspaces.
- The plugin catalog offers over 4,000 apps, plugin permissions are shared across dots, ChatGPT, ChatGPT Work and Codex, and none of the three help pages mention an analytics or search platform as of Sept. 30.

OpenAI's Dots agents can check connected sources for updates before a user makes any request — but the "always on" proactive mode revealed at DevDay this week is strictly read-only, according to the company's help pages.
The privacy and safety page 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."
OpenAI is rolling out Dots to Pro users across various markets, excluding the European Economic Area, Switzerland and the UK. Business Premium users get access in all supported ChatGPT regions. Enterprise, Edu and Healthcare workspaces can try a beta version that ships switched off by default. Access may take several days to reach an account.
Custom rules cannot loosen the restrictions
Dots support Custom Rules — per-user settings that define what a dot can do independently. These settings can make a dot's actions more or less flexible. They cannot, however, override the proactive research restrictions. Nor can the dot itself.
The unprompted mode is limited to two things: reading and taking notes. Every action the dot takes with what it finds still passes through OpenAI's usual safety checks and rules.
Scheduled checks move onto the agent
Scheduled work is not new to ChatGPT. The platform's existing Scheduled tasks feature already handles one-time and recurring jobs, watches for changes and responds to supported events from Gmail, Slack or GitHub.
A dot can run those checks across a user's connected apps and keep context between conversations. The getting-started page gives one example: a dot that reviews a calendar every morning and reports upcoming events. Both scheduled runs and proactive updates land inside the same conversation.
What the dot remembers
Dots use plugin connections the user has already set up. According to the privacy and safety page, plugin permissions are shared across dots, ChatGPT, ChatGPT Work and Codex. The launch post says the plugin catalog now offers more than 4,000 apps.
The dot also builds its own memory from those connections. Disconnecting an app stops new access but does not erase what the dot has already saved. Users currently cannot view or edit these individual memories. Resetting the dot deletes 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 connected-app examples include a calendar, Slack and a personal email account, and the getting-started page notes that available apps may vary.
Dots can browse during assigned work
The read-only ceiling applies only to unprompted activity. While working on an assignment, a dot can operate its own browser and cloud computer. Enterprise admins can enable or disable cloud browser use, cloud network access and cloud computer use individually.
Dots sit alongside the August WebMCP deployments from OpenAI, Shopify and Cloudflare, which gave agents structured methods to operate within websites. Neither the announcement nor the three support pages link Dots to search rankings or request website modifications. Dots add another agent capable of browsing a site on a user's behalf.
Why search marketers should care
Search marketing teams already check rankings, competitor pages and campaign settings on a schedule, and each check often starts from scratch. A dot can take on monitoring duties, watch the specific sources it is permitted to access and report changes. Because proactive research is read-only, a dot can detect a change overnight. Any follow-up action runs through OpenAI's action rules and safety checks.
Whether a dot fits a team's monitoring workflow depends on one thing: which platforms it can read through ChatGPT's plugins.
OpenAI says users will later be able to add more dots and scale each one's output, whether speed or monthly workload. The Enterprise release remains in beta. The open question for search and analytics teams is which SEO and analytics tools a dot can actually reach through the plugin catalog — proactive research only pulls from permitted connected sources.
via help.openai.com (Original)
More from Amara Osei
Show full bio
News editor covering industry trends and analytics at Marketing Herald.
63 articles