On Tuesday, OpenAI initiated the rollout of a new feature for ChatGPT that allows users to schedule reminders and actions within the AI chatbot. Known as Tasks, this functionality is currently available to all paid subscribers of the platform, with plans to extend access to free-tier users later this year. This move marks OpenAI’s initial attempt to enhance the agent-like capabilities of its popular chatbot.
OpenAI Introduces Tasks Feature for ChatGPT
In a series of updates shared on X (previously Twitter), OpenAI revealed that the Tasks feature is currently in beta for users of ChatGPT Plus, Teams, and Pro. The company is set to make this feature accessible to users on the free tier as well. At present, the Tasks interface can only be used on the web version of ChatGPT.
OpenAI indicated that it aims to utilize the beta phase to gather insights on user engagement with the feature, allowing for adjustments before widespread availability. It’s important to note that users are unable to utilize the Advanced Voice Mode to set tasks at this time.
Tasks manager page in ChatGPT web
Photo Credit: OpenAI
The Tasks feature essentially offers scheduling capabilities. Users can access it through the ChatGPT web interface, either by entering commands in the text field or navigating to the Tasks page from the profile menu to create a new task. These tasks may serve as one-time reminders set for specific dates and times or recur periodically. Users can also instruct the chatbot to execute a task at the designated time.
This feature allows for various practical uses, such as receiving a daily news summary in the morning or motivational messages before workout sessions. Users can set tasks using natural language prompts, for example, by specifying, “Send me a summary of the biggest technology news of the day at 7:30 PM on weekdays,” which will establish the schedule accordingly.
While the scheduling aspect of this feature may not be particularly innovative—many smartphone clock applications offer reminder options—the ability to carry out actions at a future time adds a compelling dimension. Although this function does not qualify as truly agentic, as no actions occur beyond the chatbot’s framework, it necessitates the AI to utilize memory and employ multi-step processes to generate text or images at scheduled times. Such features are typically associated with advanced AI agents.
OpenAI has been rumored to be developing its first AI agent, referred to as Operator, for a considerable period, suggesting that the introduction of Tasks may be a pivotal step toward that ambition.
