Safety Always-on agent by OpenAI

Is ChatGPT dots safe?

Short answer It depends on what you connect. OpenAI says it asks before sensitive actions, it does not see your passwords, training on your data: yes, unless you opt out; these are the vendor's own statements, not tests.

6 of 6 safety questions answered on OpenAI's own pages , Asks before sensitive actions: Yes , Sees your passwords: No , Trains on your data: Yes, unless you opt out

What OpenAI states about connecting your accounts. AideDex reports the vendor's published statements, each linked to its source; it does not test or audit them. "Not published" means we found no statement, not that the answer is no. Facts checked up to .

Can you get it
  • Status Limited (Limited rollout)
  • Where About 176 countries (Limited rollout in about 176 countries, not in 9 of the 25 countries we track (8 of them on ChatGPT Business Premium))
  • Plan $100/mo (Cheapest plan that includes it: ChatGPT Pro)
ChatGPT dots: countries, plans, capabilities →

The six questions

6 of 6 answered
  • What data can ChatGPT dots access?

    Uses the plugins you connect in ChatGPT (shared with ChatGPT, Work and Codex) within granted permissions, plus optional local computer access once you connect your computer and grant the ChatGPT app camera, microphone or screen permission. Proactive research can read permitted connected sources and save private notes, but cannot send messages, change content through plugins, or control a browser or computer. Manage plugins in the Plugins tab; disconnecting does not delete what the dot already learned. help.openai.com · checked
  • Does ChatGPT dots ask before sending, buying or deleting?

    Yes
    Changing a password or transferring money requires you to take over; deleting data or installing software may need approval each time… Full quoteChanging a password or transferring money requires you to take over; deleting data or installing software may need approval each time; Auto-review checks actions first.
    help.openai.com · checked
  • Can ChatGPT dots see my passwords?

    No Secure sign-in and saved passwords are kept out of the model's context; passwords you paste into chat or documents are not covered. openai.com · checked
  • Does ChatGPT dots use my data to train AI models?

    Yes, unless you opt out
    Personal plans: controlled by "Improve the model for everyone" (on by default per Data controls). Full quotePersonal plans: controlled by "Improve the model for everyone" (on by default per Data controls). Business/Enterprise/Edu not used by default. Not trained directly on proactive research notes.
    help.openai.com · checked
  • Where does ChatGPT dots run, and is it isolated?

    Each dot has its own sandboxed cloud computer (Linux + Chrome), isolated from other users; your computer stays separate unless you connect it. openai.com · checked
  • How long does ChatGPT dots keep my data?

    A dot keeps context from conversations and plugins for as long as you keep it; you cannot view or delete individual dot memories, only delete the whole dot. Files, Codex threads and chats it created are stored separately and are not deleted with it. help.openai.com · checked

Other things OpenAI states

2 statements
  • Human review of dot activity may occur in limited cases, including safety, even with model improvement turned off. help.openai.com · checked
  • Safety monitoring can pause or stop a dot's work if it detects a concern. openai.com · checked

More questions about ChatGPT dots and your data

1 answer
How do I reset or delete my dot and start again?
Open your dot's profile, select the more options menu (•••), then Reset and confirm. Reset deletes the dot with its conversations, saved memories and scheduled tasks. You then return to a new ChatGPT chat and can create a new dot in the mobile app, desktop app or desktop web. help.openai.com · checked

Compare and read on

All alternatives to ChatGPT dots · Is it safe to connect Gmail to an AI agent? What each vendor says it does · How facts are checked

Sources for this page

4 pages
  1. help.openai.com/en/articles/20001529-dots-privacy-security-and-safety-faqs
  2. openai.com/index/how-we-build-safety-security-and-privacy-into-dots/
  3. openai.com/index/introducing-dots/
  4. help.openai.com/en/articles/20001530-getting-started-with-your-dot

Spotted something wrong or out of date? Report an error · Corrections log