OpenAI has introduced a new feature called Tasks, which allows ChatGPT to create reminders and carry out scheduled actions. The feature is already available in beta and is an attempt to turn the chatbot into something more, bringing it closer to the functionality of such virtual assistants as Google Assistant or Siri.
Image source: Andrew Neel / Unsplash
Tasks allows you to give ChatGPT a task, exactly what needs to be done and when. For example, you can ask to be sent a daily weather forecast at a certain time, remind you about the expiration date of your passport, or even prepare a fairy tale to tell your children before bed, explains The Verge.
To use the feature, subscribers need to select the “4o with scheduled tasks” model from the ChatGPT menu, then simply formulate their request for what needs to be done and indicate the execution time. Additionally, Tasks can recommend tasks based on previous conversations, but the user must manually approve each of these recommendations. You can manage tasks both in the chats themselves and in the new Tasks section, available via the web version. Once tasks are completed, users will be able to receive notifications on all their devices. It is important to note that up to 10 active tasks can be running simultaneously.
The innovation is already available in beta version for subscribers of Plus, Team and Pro tariffs. OpenAI hasn’t yet said whether the feature will be available for free use, but Tasks will presumably remain part of its premium plans, which cost $20 or $200 per month.
Let us remind you that previously the chatbot worked exclusively in real time, responding to requests as they were received. However, OpenAI decided to slightly expand the scope of ChatGPT, turning it into a tool for long-term planning. Moreover, new developments are on the way, such as the autonomous agent Operator, which will be able to control computers, and the Caterpillar project, which can integrate with Tasks for information search, data analysis, site navigation and work with documents.
It is noted that despite the prospects, the function still has unresolved issues. For example, how reliably will Tasks cope with tasks, especially when it comes to important and urgent information. And the launch in beta indicates that OpenAI is still only testing this function to collect feedback from users, without guaranteeing 100% reliability.