OpenAI quietly shipped a capability on June 17, 2026 that changes how most people will think about ChatGPT: scheduled tasks. Instead of waiting for you to ask a question, ChatGPT can now run work on a schedule and come back with results — while you’re in a meeting, asleep, or not even thinking about it.

This is not a minor release note. It’s a signal about where the whole AI industry is heading.

What Actually Launched

The update gives ChatGPT users a dedicated Scheduled page, accessible from the sidebar on both web and mobile. From there you can set tasks to run at specific times or loose windows (morning, afternoon, evening). You can also set up monitoring tasks — where ChatGPT periodically checks the web or connected apps for changes and notifies you only when something meaningful happens.

Think of it as giving ChatGPT an alarm clock, a to-do list, and a watch service all at once.

Available plans include Go, Plus, Pro, Business, and Enterprise. Enterprise accounts can run up to 15 active tasks simultaneously. There’s a cap of one run per hour per task.

OpenAI is also sunsetting Pulse, the daily summary feature it launched last year. Pro users get a 14-day wind-down period, and the suggested alternative is to simply schedule a daily briefing as a task.

Why This Matters More Than It Sounds

For most of the past three years, AI tools have been reactive. You prompt, they respond. Useful, but fundamentally still dependent on a human initiating every step.

Scheduled tasks break that pattern. ChatGPT can now check whether a competitor updated their pricing, monitor a news topic and surface what changed, run a recurring analysis at the start of each week, or send a reminder tied to specific conditions — all without waiting for you to log in and ask.

For a business owner, that shift has real value. How many recurring tasks in your week are really just “check this, summarize it, tell me if something changed”? Probably more than you think.

The Broader Pattern: AI Is Moving Into the Background

This launch sits inside a larger trend that business leaders need to understand. AI is moving from a tool you use to infrastructure that runs underneath your work.

The most consequential AI deployments over the next 12 months will not be the ones where you chat with an AI assistant. They will be the ones where an AI process runs continuously in the background, monitoring, alerting, handling routine tasks, and escalating only when a human decision is actually needed.

OpenAI is shipping that direction into a consumer product that already has over a billion monthly users. That matters because it normalizes autonomous AI workflows for a huge slice of the business world — including the customers you want to keep.

What This Means for Business

For teams using ChatGPT Business or Enterprise: Start mapping which recurring research, monitoring, or prep tasks could now be delegated. Morning briefings, weekly competitive checks, client update summaries — these are immediate candidates.

For data and analytics teams: Scheduled monitoring of dashboards or data sources is a practical use case, though you will still want purpose-built data tooling for anything serious. ChatGPT is a generalist; scheduled tasks in a proper data environment are a different conversation.

For business owners evaluating AI strategy: The fact that OpenAI’s consumer tool now runs autonomous tasks is useful context — but consumer tools are not enterprise AI implementations. The gap between scheduling a ChatGPT reminder and deploying an actual AI agent workforce (with governance, reliability, and integration into real business systems) remains large.

The useful question is not “should I use ChatGPT scheduled tasks?” The useful question is “what does autonomous AI running in the background of my business actually look like at scale?”

The Honest Limitation

Scheduled tasks in ChatGPT are a useful feature. They are not an agent workforce. They lack persistent memory across tasks (beyond what you configure), they operate within a single tool rather than across your business systems, and the 15-task limit means enterprise automation at scale needs a different approach.

Think of it like a smart alarm clock for your AI assistant — a meaningful improvement over having to remember to ask, but not a replacement for a proper AI operations layer.

For businesses serious about deploying autonomous AI across their workflows, the architecture needs to be built deliberately: agents grounded in your actual business processes, data, and rules — not a consumer app running in the background.

---

OpenAI Scheduled Tasks is available now on web and mobile for Plus, Pro, Business, and Enterprise users. The feature is rolling out globally.

Enterprise DNA helps businesses build proper AI agent workforces — not just experiment with consumer tools. Book a discovery call to talk about what autonomous AI actually looks like inside your operations.