…বাংলা
Barlekha Today logo
All the big stories — in your language, updated every 15 minutes
Home › Technology › News
Technology

ChatGPT's AI Assistant 'Dot' Can Now Be Given Tasks by Text Message

Staff Reporter
ChatGPT's AI Assistant 'Dot' Can Now Be Given Tasks by Text Message
Photo: Collected

OpenAI, the company behind ChatGPT, has launched a feature that lets users give work instructions to its artificial intelligence personal assistant 'Dot' by sending it a direct message. Under the new system, users can explain the task they want done not only through text messages but also by sending photos and voice memos.

Dot will not only receive instructions; it will also be able to message the user on its own to report the progress of a task given earlier or to share necessary information. Initially, however, the feature is being rolled out experimentally for ChatGPT Pro subscribers in the United States.

When the Dot AI agent was launched in late September, OpenAI had said that a messaging option for communicating with the agent would be added in the future. The implementation of that plan has now begun.

Dot has been built not merely as a chatbot that answers questions, but as an AI assistant that performs various tasks on the user's behalf and monitors specific subjects to report useful information. With the new messaging feature, users will no longer need to open the ChatGPT app every time; they can communicate with the assistant simply by sending a message.

Under the new system, iPhone users will be able to reach Dot through Apple's iMessage and Android users through RCS messaging. To enable the feature, users must go to Dot's profile in the ChatGPT app and select the 'Texting' option, then make sure iMessage or RCS is active on their phone.

OpenAI says it is working with a third-party company called 'Link' to provide the technology for exchanging the messages.

On security, OpenAI says the messaging system has been designed so that a user can message only their own Dot, and Dot can message only that user. While the feature is on, Dot will have no opportunity to read the user's other personal messages. The company says it used its own advanced AI models as well as third-party security experts to verify the system's safety.

The same memory and plugins available in the ChatGPT app can be used in Dot's messaging. As a result, instructions given earlier in the app and stored relevant information will also be usable in message conversations.

For now, however, the feature remains limited to Pro subscribers in the United States and at an experimental stage. OpenAI has given no specific timeline for when users in other countries, or free users, will get it.

More in Technology