OpenAI tonight presented another big news in the field of artificial intelligence. It's called Dots and it's a so-called "always-on"AI agents that are supposed to be able to work on your tasks virtually continuously. OpenAI He describes them as agents "designed to handle everything," adding that the new feature is aimed directly at the rival Muse from Meta.
Dots use the model GPT-6 Astra, with each agent having their own clouda computer with a web browser. So it's not just another chatbot, to which you ask a question and wait for an answer. The agent can be a long-term
OpenAI tonight introduced another big news in the field of artificial intelligence. It's called Dots and it's a so-called "always-on"AI agents that are supposed to be able to work on your tasks virtually continuously. OpenAI He describes them as agents "designed to handle everything," adding that the new feature is aimed directly at the rival Muse from Meta.
Dots use the model GPT-6 Astra, with each agent having their own clouda computer with a web browser. So it's not just another chata bot that you ask a question to and wait for an answer. The agent can work on the given goals over a long period of time, conduct background research and gradually learn from your feedback. So the longer you work with it, the better it should know your preferences and way of working.
You might be interested in
One agent across ChatGPT, Slack and Teams
What is particularly interesting for me personally is the connection with other services. Dots is, according to OpenAI they can connect to more than 4000 applications and you can communicate with them not only via ChatGPT, but also via Slack or Microsoft Teams. In the future, it is even planned to be possible to communicate via classic text messages. The agent will be able to transfer context between individual communication channels.
OpenAI Moreover, it did not forget about permissions. Dots follows the permissions set in ChatGPT and users can create their own rules specifying which tasks the agent must first request approval for before performing the action.
Starting today, users can create their own main agent, which can also be named. Later, it has OpenAI enable the creation of entire Dots teams, allowing multiple specialized agents to collaborate on a single project. The new feature is now gradually being made available to tariff users Pro, Business Premium and Enterprise in select markets, with additional users to follow at a later date. Eligible users will receive one agent at no additional charge. Conversations with them do not count towards limits ChatGPT, but the tasks themselves use the limits of the respective tariff.
Dots can be set up via a desktop application ChatGPT and then access them from the mobile application. And if the whole solution works as it should OpenAI promises, I'm not afraid to say that it moves forward ChatGPT again a bit further from the classic chatAI towards a digital assistant that can work on assigned tasks without you having to sit next to it all the time. Moreover, such innovations fundamentally "untie the hands" of those who know how to use them to the fullest. Therefore, for me, this is a really nice upgrade.
They will work on your goals, conduct background research, and gradually learn from your feedback. The longer you work with them, the better they will know your preferences and way of working.
Gallery
One agent across ChatGPT, Slack and Teams
What is particularly interesting for me personally is the connection with other services. Dots is, according to OpenAI they can connect to more than 4000 applications and you can communicate with them not only via ChatGPT, but also via Slack or Microsoft Teams. In the future, it is even planned to be possible to communicate via classic text messages. The agent will be able to transfer context between individual communication channels.
OpenAI Moreover, it did not forget about permissions. Dots follows the permissions set in ChatGPT and users can create their own rules specifying which tasks the agent must first request approval for before performing the action.
Starting today, users can create their own main agent, which can also be named. Later, it has OpenAI enable the creation of entire Dots teams, allowing multiple specialized agents to collaborate on a single project. The new feature is now gradually being made available to tariff users Pro, Business Premium and Enterprise in select markets, with additional users to follow at a later date. Eligible users will receive one agent at no additional charge. Conversations with them do not count towards limits ChatGPT, but the tasks themselves use the limits of the respective tariff.
Dots can be set up via a desktop application ChatGPT and then access them from the mobile application. And if the whole solution works as it should OpenAI promises, I'm not afraid to say that it moves forward ChatGPT again a bit further from the classic chatAI towards a digital assistant that can work on assigned tasks without you having to sit next to it all the time. Moreover, such innovations fundamentally "untie the hands" of those who know how to use them to the fullest. Therefore, for me, this is a really nice upgrade.
You might be interested in
