OpenAI has introduced Operator, an AI agent capable of performing web-based tasks autonomously, marking a significant leap in the company’s advancement of AI technology. CEO Sam Altman described the launch as a pivotal moment for 2025, underscoring the importance of AI agents in the future of technology.
Operator, a general-purpose AI tool, is designed to handle tasks such as booking travel, making restaurant reservations, and online shopping. Initially available to US users through a $200 Pro subscription, OpenAI plans to expand access to Plus, Team, and Enterprise tiers, and eventually to more regions, although European users will face delays.
Also Read: OpenAI Launches Sora, a Text-to-Video Model, for ChatGPT Plus and Pro Subscribers
The AI agent operates by controlling a dedicated web browser, performing actions like navigating menus and filling out forms. Users can monitor the process and provide confirmation before tasks with external consequences, such as purchases or emails, are finalized. Operator uses OpenAI’s Computer-Using Agent (CUA) model, which blends GPT-4’s vision capabilities with advanced reasoning to tackle a variety of online tasks without relying on APIs.
Operator’s current capabilities span across shopping, delivery, dining, and travel, with partnerships in place with companies like DoorDash, Uber, and eBay. Despite its powerful capabilities, OpenAI admits the tool has limitations, particularly with specialized tasks or complex web interfaces.
Security measures are a key focus for OpenAI, as users are required to manually input sensitive information, and certain tasks, like banking transactions, require user oversight. Restrictions also exist on the number of tasks Operator can perform and the types of tasks it can execute, ensuring safe use.
The launch of Operator positions OpenAI alongside competitors like Google and Anthropic, marking a major milestone in the development of AI agents capable of autonomous task completion.
