COR Brief
Join Beta
AI Tools›Agents & Automation›OpenAI Operator
Agents & Automation

OpenAI Operator

OpenAI Operator is an AI agent developed by OpenAI that automates browser-based tasks by interacting directly with web graphical user interfaces. It executes tasks such as filling out forms, ordering groceries, or creating memes by navigating websites and performing clicks and text entries as a human user would, without relying on APIs. Users provide natural language instructions, and the agent operates in a remote browser environment, allowing users to take control for sensitive steps like logins or CAPTCHAs. The system includes safety features such as declining high-risk tasks and watch mode supervision for sensitive sites. Initially powered by the Computer-Using Agent (CUA) model combining GPT-4o vision and reinforcement learning, OpenAI Operator transitioned to an o3-based model and was fully integrated into ChatGPT as "agent mode" by mid-2025. The standalone site operator.chatgpt.com is scheduled to sunset following this integration. Access is currently limited to ChatGPT Pro users in the U.S., with plans to expand availability. Additionally, a CUA API is available in research preview for select developers to integrate automation into applications or workflows.

Updated Jan 24, 2026subscription
Visit OpenAI Operator ↗Visual Guide
Overview

OpenAI Operator automates web tasks by controlling browsers through natural language commands within ChatGPT's agent mode.

Pricing
Included in Pro subscription (U.S. only)
Category
Agents & Automation
Company
Visual Guide
Interactive PresentationOpen Fullscreen ↗
Key Features
01
Executes browser tasks such as form filling and ordering by interpreting user prompts and interacting with web GUIs like a human.
02
Allows users to manually intervene during sensitive steps like logins, payments, or CAPTCHAs when prompted by the agent.
03
Supports custom instructions for specific websites and saving prompts for repeated tasks to improve efficiency.
04
Runs multiple browser automation tasks simultaneously across separate conversations.
05
Includes watch mode for monitoring sensitive sites and declines tasks involving high-risk actions per usage policies.
Real-World Use Cases

Automated Grocery Ordering

User instructs Operator to order groceries on Instacart via natural language.

Form Filling for Travel Booking

User requests booking a flight with airline preferences on Booking.com.

Quick Start
1
Access OpenAI Operator
Use ChatGPT and select 'agent mode' from the composer dropdown (requires ChatGPT Pro subscription, U.S. only).
2
Enter Task Description
Type a natural language instruction describing the web task you want automated, such as 'order groceries on Instacart.'
3
Monitor and Take Control
Watch the remote browser session and take control when prompted for sensitive actions like logins or CAPTCHAs.
4
Customize and Save Prompts
Add custom instructions for specific sites or save prompts for tasks you perform repeatedly.
5
Developer API Access
If eligible, install the Agents SDK (pip install openai-agents) to use the CUA API in research preview for integration.
📊

Strategic Context for OpenAI Operator

Get weekly analysis on market dynamics, competitive positioning, and implementation ROI frameworks with AI Intelligence briefings.

Try Intelligence Free →
7 days free · No credit card
Pricing
Model: subscription
ChatGPT Pro
Included in Pro subscription (U.S. only)
  • Access to OpenAI Operator via ChatGPT agent mode
  • Multiple simultaneous tasks
  • Custom instructions and saved prompts

Currently available to ChatGPT Pro users in the U.S. with plans to expand to Plus, Team, and Enterprise tiers. The CUA API is in research preview for select developers with pay-as-you-go pricing not publicly detailed.

Assessment
Strengths
  • Automates diverse web tasks by interacting with web GUIs without requiring APIs.
  • Supports multiple parallel tasks and saved prompts for repeated automation.
  • Includes safety features such as watch mode and refusal of high-risk tasks.
  • Integrated into ChatGPT for seamless access post-July 2025.
  • Offers a developer preview API for integration into applications and workflows.
Limitations
  • Access limited to U.S. ChatGPT Pro users initially, with no availability in EEA or Switzerland yet.
  • Technical issues reported, including failures in user takeover due to data channel errors.
  • Declines many sensitive tasks, limiting automation scope in some scenarios.
Alternatives
Anthropic Claude with Computer Use→xAI Grok→Browserless / Puppeteer→MultiOn→