OpenAI launches ChatGPT Dots, AI assistants that work around the clock on your behalf

OpenAI wants its AI to stop waiting for commands to act. At the opening keynote of the Dev Days, Sam Altman, the head of OpenAI, unveiled “Dots,” AI agents integrated into ChatGPT.The CEO describes them as “remarkably capable” and “always-on”, built to handle “anything you can imagine”.

Introducing dots, powered by GPT-6 Astra.

Remarkably capable, always-on agents built to handle everything. pic.twitter.com/inY5BZj3Kz

— OpenAI (@OpenAI) September 29, 2026

In practice, Dots work around the clock on the user’s behalf. Sam Altman promises an AI “that knows what’s happening” and “what matters to you”. “You don’t have to tell it everything”, he added. According to him, today’s virtual assistants, which can book a restaurant or buy a plane ticket, are “trivial compared with what the technology can really do”.

An assistant that manages your entire life

During the demonstration, a Dot agent, nicknamed “Alfred” by its user, completed one mission after another. It built a website, updated a presentation for a board of directors, and found a replacement caterer for a wedding. It even moved a meeting that clashed with the show put on by the user’s daughter. Rest assured, the AI agent still asks for the user’s approval before making important decisions. Dots can also tackle complex computing tasks.On stage, Sam Altman notably cited the migration of an application away from an old interface that is about to be discontinued.

To achieve this, each Dot has “its own computer in the cloud” and its browser. It respects the user’s access rights and works with more than 4,000 applications. In plain terms, OpenAI is taking the idea of ChatGPT Work, its agent launched this summer, much further here. Soon, it will also be possible to text your Dot via SMS, or even call it. Sam Altman clarified that you can start with a single Dot and then, “in the future”, have a whole team. Soon, the user will be able to head a team of several AIs to manage t

On the security front, OpenAI says that Dots are based on “Astra, our best-aligned model”. The user can also set limits on the use of applications and the computer. According to Sam Altman, this makes it possible to entrust your Dot with “as many responsibilities as you feel comfortable giving it”. These safeguards aim to address the concerns raised by ChatGPT’s various missteps in recent weeks.

ChatGPT Spaces, the shared workspace

OpenAI also took the opportunity to launch ChatGPT Spaces. For Sam Altman, “almost all productivity software was not designed for AI and humans to work well together”.That's why OpenAI launched Spaces, a place where “you, your colleagues and your agents” collaborate. Each space brings together pages and files, much like a shared drive. You can write texts, conduct research, or generate images and data in it.

In fact, you just need to mention a Dot in a comment for it to get to work. OpenAI has no intention of stopping there. The startup will soon offer a presentation tool that agents will be able to read and edit easily, under the supervision of the person they are talking to. Dots and Spaces are available today for ChatGPT Pro, Business Premium and Enterprise subscribers. Dot is included in the ChatGPT subscription, and its conversations with the AI do not count against the plans’ usage quota.

Bad news, however, for European users. Dots and ChatGPT Spaces are not available in Europe at the moment.

👉🏻 Follow tech news in real time: add 01net to your sources on Google, and subscribe to our WhatsApp channel.