The agent that works while you sleep is a different product category. openai's dots runs in its own…
the agent that works while you sleep is a different product category. openai's dots runs in its own cloud vm with a browser and keeps writing code and pushing projects while you're offline.
trust moved from the answer to the unsupervised run.
Context
OpenAI's launch post for dots describes always-on agents that work in their own cloud computer with their own browser and the apps a user has connected. It says a dot 'can take a project and run with it' while handling several others, and gives a developer example where a dot builds and tests fixes and brings complete pull requests to review. Dots are rolling out on Pro, Business Premium and Enterprise plans in eligible markets.
The same post says that when you are not actively working with a dot, it does 'proactive research' using connected apps through tools restricted to read-only, which cannot send messages, change app content or control your browser or computer. Search Engine Journal (September 30, 2026) quotes OpenAI's help page to the same effect and says the mode that runs without a prompt is limited to reading and taking notes.
The post also says dots can message the user with progress, questions or decisions, and that safety monitoring can pause or stop a dot's work.
The note says OpenAI's dots 'keeps writing code and pushing projects while you're offline'. OpenAI's own page describes dots working on projects in a cloud computer, but the part that runs without any prompt is stated as read-only research. Whether a dot writes code and pushes work unattended after being assigned a project is not spelled out in the pages read, and the developer example speaks of PRs brought to the user for review. So 'pushing projects' while offline is unsupported here, not refuted.
The rollout covers Pro, Business Premium and Enterprise in eligible markets, so the product is not available to every user yet.
'Trust moved from the answer to the unsupervised run' is the author's opinion. The sources describe permission settings and read-only limits as the controls on unsupervised work.
Related work
- Introducing dots (OpenAI) ↗Primary source for what dots do, where they run and the rollout.
- OpenAI Dots Run Read-Only Research When Nobody Is Asking (Search Engine Journal) ↗Source for the read-only limit on work done without a prompt.
- OpenAI launches Dots, its bubbly agentic avatar (TechCrunch) ↗Launch coverage from September 29, 2026.
Watch next
- Read OpenAI's dots help pages for what a dot may do on its own after a task is assigned. Check the permission levels set by Custom Rules.
Sources
- Introducing dots (OpenAI)openai.com
- OpenAI Dots Run Read-Only Research When Nobody Is Asking (Search Engine Journal)searchenginejournal.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 8 October 2026 at 22:40 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →