OpenAI is shutting down ChatGPT Pulse, its proactive content briefing feature. The company plans to replace it with a more versatile Scheduled Tasks system. The move highlights the tech giant’s focus on turning ChatGPT into an action-oriented AI assistant rather than a tool that simply delivers information.
Pulse was designed to provide customers with personalized updates according to their preferences, conversations, and connected applications. It provided users with curated briefs without requiring any searching.
Although Pulse demonstrated ChatGPT's proactive capabilities, it had a limited number of use cases. OpenAI is moving toward developing more versatile tools that will give people greater control. In an X post, OpenAI announced, “We’re also sunsetting Pulse in the next 14 days, but you can keep getting daily updates—use scheduled tasks to get a daily briefing based on your interests, past chats, and connected apps.”
Scheduled Tasks allow users to control ChatGPT to perform tasks at specific times. This means that instead of getting some existing data, you can choose what you want to know, when, and how often. This includes reminders, reports, updates, and monitoring requests. Customers can also edit, manage, and cancel tasks in a special interface.
The shift aligns with OpenAI's overall goal of developing agentive AI models that can offer proactive support to users for everyday tasks. These AI models can perform:
Regular tasks
Stay updated on the latest trends
Provide updates when required without being prompted
The introduction of Scheduled Tasks aligns with OpenAI’s broader vision for agentic AI. This feature allows users to automate tasks of their choice while maintaining control over what ChatGPT does on their behalf.
One of the biggest differences between Pulse and Scheduled Tasks is user control. Pulse can generate updates based on available context. However, Scheduled Tasks allows people to define specific outcomes and schedules.
Also read: ChatGPT Suicide Allegations Intensify Calls for Stronger AI Safeguards