The news
OpenAI Dots, launched on September 29, 2026, at the company's DevDay conference, are always-on agents inside ChatGPT that run on their own cloud computers and keep working toward goals without step-by-step prompting, giving software ongoing access to the apps where work happens.
The agents run on GPT-6 Astra, OpenAI's flagship model. Each Dot gets its own cloud computer and browser, which users can open at any time to see what it is doing, according to SiliconANGLE and VentureBeat. Through OpenAI's plugins, a Dot can reach more than 4,000 apps, and with permission it can also work directly on its owner's laptop.
Users can reach a Dot in ChatGPT on desktop, web or mobile, and in Slack and Microsoft Teams, with context carried between channels. TechCrunch reported that text message support is coming, and VentureBeat reported that CEO Sam Altman said Dots will also come to other messaging apps and to the phone as an audio model. OpenAI said it eventually envisions teams of Dots working together for one person.
OpenAI has been using the agents internally, SiliconANGLE reported: when a bug is reported in the company's Slack, agents start investigating, and new designs are turned into working apps. In one outside test, an agent noticed its owner had forgotten to bill a publication, prepared the invoice and waited for his approval before sending it.
Dots began rolling out on September 29 to ChatGPT Pro and Business Premium subscribers, though some accounts may wait several days. Pro subscribers in the European Economic Area, Switzerland and the U.K. are excluded. Enterprise, Edu and Healthcare customers can try a beta once a workspace administrator enables it. One Dot comes with Pro or Business Premium at no extra cost, and usage will not count against plan allowances for the first month, SiliconANGLE reported, citing OpenAI's help documentation.
OpenAI built in limits. When a Dot has no assigned task, it does what OpenAI calls proactive research, using read-only tools that cannot send messages, change content or control the user's computer. Actions that could touch accounts or share information must clear a check called auto-review. Custom Rules let organizations allow, require approval for or block actions, an Activity View shows background work, and some sensitive steps, including changing passwords, stay with the human. OpenAI tells users to review consequential work because the agents can still make mistakes.
The numbers
- Apps reachable through OpenAI plugins
- More than 4,000
- Dots included with Pro or Business Premium
- 1, at no extra cost
- Period when Dot usage does not count against plan allowances
- First month
- Model powering Dots
- GPT-6 Astra
- Launch date
- September 29, 2026
Why CEOs should care
For CISOs, the change is standing access. A chatbot answers when asked; a Dot keeps working inside whatever apps it has been connected to. Specialist Dots, which OpenAI is piloting, get their own organizational identity, credentials and access to company systems, VentureBeat reported. Treat each agent like a new machine identity: decide which apps it may reach, which actions need approval under Custom Rules, who watches the Activity View, and whether auto-review decisions are logged somewhere your security team can see.
For CIOs, the rollout path is the risk. Pro and Business Premium users can switch Dots on themselves, while Enterprise, Edu and Healthcare workspaces get the beta only when an administrator enables it. If employees pay for personal Pro plans, they may connect work accounts to an agent that sits outside company controls. Set a written policy now on connecting work apps, and give laptop access, which Dots can receive with permission, an explicit yes or no.
For CFOs and boards, the bill is still unknown. OpenAI has not disclosed prices for additional Dots, extra speed or monthly capacity, or specialist Dots, and the free usage window ends after the first month. Run small pilots, track how often a person has to step in, and hold off on multi-agent commitments until pricing is published. Boards should ask who is accountable when an always-on agent acts for the company.
The bigger picture
Dots put OpenAI into a fast-forming race for persistent personal agents. Meta Platforms (META) released its Muse personal agent in the U.S. on September 8; Muse also runs on a dedicated cloud computer, and Meta says most use is free, with subscriptions for heavier users. VentureBeat also pointed to SpaceXAI's Grok Bot, launched in August, which gives agents their own computers and keeps them running around the clock. TechCrunch noted that much of what Dots do was already possible through Codex and similar agent tools; the new part is packaging it as an independent, always-running assistant.
Microsoft (MSFT) is working with OpenAI so specialist Dots can be governed through Microsoft Agent 365 security and management controls. VentureBeat argued that companies may end up treating agents less like software licenses and more like machine identities that need onboarding, credentials, monitoring and offboarding.
What’s next
Watch for OpenAI to publish prices for additional Dots and capacity once the first month ends, for wider specialist Dot pilots, in which OpenAI engineers help customers define each agent's duties, tools and approvals, and for the Agent 365 integration. Text messaging and phone access are also planned, according to OpenAI.
What “Fact-checked” means
Fact-checking means testing a story’s facts against the evidence before it is published. This story went through at least two separate checks before this version was published.
- What we checked
- Its names, figures, dates, job titles, quotes and who said what were checked against the story’s sources, including its main source where it could be opened. The headline was checked for accuracy and overstatement.
- How
- A first check reviewed the whole story. If it passed, a second, skeptical check went back to the sources to look for mistakes in the most important facts. If a check flagged the story, it was edited to fix the problems found, and a separate re-check then reviewed the whole story again.
- Who
- The checks are made by our newsroom, as steps kept separate from the writing, under rules set by our editor, Hussein Mukhtar. A story the checks still flag is held for the editor, who decides whether it is fixed, published or dropped.
- If something is wrong
- “Fact-checked” does not mean error-free. If a material error is found after publication, we correct the story and add a note saying what changed. Report an error
