OpenAI Tests Email Addresses for Dots AI Assistants
The company is reportedly giving its always-on dots assistants their own email accounts, drawing privacy and security concerns from experts.
Updated on Nov. 18, 2026 in Artificial Intelligence

Live Poll
Would you trust an AI assistant with access to your email account?
OpenAI is reportedly testing giving its dots AI assistants their own email addresses, according to a November 17, 2026 post on X by software engineer Samir Ahmed. Privacy and cybersecurity experts have raised concerns about the always-on assistants, which OpenAI rolled out on September 26, 2026.
Why it matters
Email access would give dots more autonomy, which raises the stakes if the assistants make mistakes or are compromised. A survey of 900 tech leaders by Gravitee found that exposing company data to the wrong people is the most widely reported AI agent incident.
Dots are powered by GPT-6 Astra and run on their own cloud computer, learning from user feedback over time. OpenAI says built-in monitoring can pause or stop a dot's work when a safety concern is detected, but the extraction does not specify how often this has been triggered.
The players
OpenAI
An artificial intelligence research company that developed the GPT model family and launched the dots assistants.
Samir Ahmed
A software engineer at OpenAI who publicly described the email testing for dots on X.
Gravitee
A company that surveyed 900 tech leaders on AI agent incidents and found data exposure to be the most commonly reported problem.
The details
Samir Ahmed said users can CC a dot on an email to schedule meetings or forward a receipt to negotiate refunds. Dots use auto-review to check actions that could affect users' accounts or share information before acting.
Timeline
September 26, 2026: OpenAI rolled out its always-on dots AI assistants.
November 17, 2026: Samir Ahmed posted on X about testing dot email addresses.
The Tech Race
OpenAI's dots launch arrived weeks after Meta rolled out its Muse assistant, which encourages users to store personal data about loved ones, including people who do not use Muse. Both companies are racing to make always-on AI assistants more autonomous while navigating the privacy questions that come with them.
If dots gain email access, users who CC them on messages could have the assistant schedule meetings or pursue refunds on their behalf. Users should review what permissions any assistant holds over their accounts and shared information before relying on it.
The takeaway
Before giving any AI assistant access to your email or accounts, check what it can read, send and share on your behalf. Limiting those permissions to what you actually need lowers the risk if something goes wrong.
Further reading
For broader context on how these assistants are changing, see the Artificial Intelligence section.
Live Poll
Would you trust an AI assistant with access to your email account?



