OpenAI is bringing scheduled tasks to ChatGPT’s free tier, giving non-paying users a limited but practical way to have the chatbot run prompts automatically. Free accounts can create up to three active tasks, marking a notable expansion of a feature that had previously been centered on paid plans.

For everyday ChatGPT users, the appeal is simple: instead of reopening the app and typing the same request each morning, they can set a prompt once and let ChatGPT handle it later. A task might deliver a daily briefing, remind someone about an upcoming event, or produce a recurring roundup based on a saved instruction.

The free version comes with clear guardrails. Users can have only three active tasks at a time, and recurring free-tier tasks can run no more than once a day. OpenAI also says free users work with flexible delivery windows such as morning, afternoon, or night, rather than paid-tier features like precise delivery times or hourly schedules.

That makes the rollout feel less like full-blown automation and more like a useful introduction to it. Three slots are enough for the kinds of routines many people actually repeat: a morning news digest, a daily language exercise, and an evening reminder to plan the next day. But it is not designed for users who want ChatGPT constantly checking for updates, running workflow-heavy research, or functioning as a personal operations assistant.

The feature sits somewhere between a reminder app and a lightweight AI assistant. Traditional reminders tell users to do something. A ChatGPT task can potentially do a piece of the work itself, provided the request is clear enough. Someone could ask for a simple explainer of the day’s biggest AI news, a weekday vocabulary lesson, or a recurring list of ideas for meal planning. The value is not merely that ChatGPT remembers the schedule – it is that the output can be shaped by the user’s original prompt.

OpenAI’s latest update also expands scheduled tasks for paid users in a more ambitious direction. ChatGPT Work users on Plus and Pro can set tasks to respond when something changes in connected services including Slack, Gmail, and GitHub, rather than relying solely on a clock-based schedule. That distinction matters: free users are getting scheduled automation, while higher tiers are beginning to move toward event-driven work assistance.

The plan limits show how OpenAI is positioning the feature. Free and Go users can keep three active tasks, Plus users can have five, Business and Education users can have 10, and Pro and Enterprise users can have 15. In other words, the free rollout is generous enough to make the feature useful, but restrictive enough to preserve scheduled tasks as one of the practical reasons to upgrade.

There are also a few operational details worth knowing before treating tasks as a set-it-and-forget-it system. If a user reaches the three-task cap, they will need to pause, delete, or wait for an existing one-time task to finish before making another. OpenAI notes that model availability, usage limits, and certain capabilities can vary depending on the task and the user’s plan. The safe approach is to use tasks for low-stakes recurring help, not for time-critical work that cannot tolerate a missed or imperfect response.

For OpenAI, opening the feature to free users is as much about habit-building as it is about feature parity. Scheduled tasks encourage people to see ChatGPT not only as a place to ask one-off questions, but as a service that can return on its own with something useful. That shift is important in a crowded AI market, where the long-term battle is increasingly about whether assistants become part of a user’s daily routine rather than just an occasional search alternative.

The three-task limit may sound modest, but it could be enough to change how free users interact with ChatGPT. The most useful AI features are often not the flashiest ones. They are the ones that quietly remove a repeated step from the day.