ChatGPT scheduled tasks page: controls, limits, and rollout

OpenAI on Jun 20, 2026 added a new Scheduled page to ChatGPT's sidebar that centralizes recurring tasks and task controls. The feature is available to Plus, Pro, Business, and Enterprise users, and the previous Pulse feature is being retired and folded into scheduled tasks.

What does ChatGPT's Scheduled page include?

The Scheduled page collects all active recurring tasks in one place and lets users view, pause, edit, or delete tasks directly from the sidebar. Research-style tasks can search the web and connected apps and will send alerts only when something actually changes; users can schedule tasks for specific times or parts of the day: morning, afternoon, or evening.

Details in the interface focus on persistent monitoring rather than one-off prompts. The UI groups recurring jobs so users no longer need to hunt through chat history to see what is actively running. That grouping is also where users can manage task cadence and stop notifications without deleting the underlying task.

Who can use Scheduled tasks and what limits apply?

The feature is available to Plus, Pro, Business, and Enterprise users, with the number of active tasks varying by plan. Tasks can run at most once per hour and will pause automatically if the user goes inactive, and OpenAI says all tasks are now faster and more reliable.

Availability is tiered rather than universal, and OpenAI ties active-task allowances to plan level. The hourly run cap and the automatic pause on inactivity place hard constraints on how aggressively the system can poll external sources or run background checks. For users who relied on the older Pulse feature, OpenAI is folding that functionality into this Scheduled experience rather than keeping both parallel systems.

Why it matters

Centralizing recurring tasks and folding Pulse into one Scheduled page tightens the product toward background monitoring and automated alerts. For users who want ongoing checks across the web and connected apps, the new controls remove friction around managing and stopping those checks. The hourly run limit and inactivity pause curb resource use and notification noise, shaping how teams and power users will design monitoring workflows.

OpenAI's claim that tasks are "faster and more reliable" signals the company expects the feature to be used continuously rather than sporadically. That expectation raises questions about how plans will constrain volume and which use cases will remain practical within the stated limits.

What to watch

Monitor how the per-plan limits on the number of active tasks are enforced and whether OpenAI adjusts the “at most once per hour” run cap. Also watch whether folding Pulse into Scheduled removes any legacy behaviors users relied on and how the change affects alert volume from research tasks that query web and connected apps.