OpenAI Introduces Operator

Something new is brewing in the world of AI, and it’s about to change how we navigate the digital world.

Fez – OpenAI, the company behind ChatGPT, has announced the launch of Operator, an AI tool designed to carry out online tasks like planning vacations, booking restaurants, or shopping.

This development marks a significant step forward in the race to create more advanced AI assistants.

Described by OpenAI as one of its first “agents,” the tool operates independently using its own browser.

It can navigate web pages, scroll, click buttons, and fill out text fields—just as a person would.

Currently, this feature is available only to ChatGPT’s professional subscribers to gather feedback and refine its capabilities.

“Operator is one of our initial agents, designed to perform tasks autonomously. You assign it a task, and it executes it,” the California-based company stated in a press release.

Since the launch of ChatGPT in late 2022, the rapid growth of generative AI has intensified competition among tech giants.

Companies are racing to create AI tools capable of composing messages, answering questions, generating images, and more.

The ultimate goal is to develop AI agents that act as all-in-one assistants—ready to manage emails, complete online shopping, and handle various other tasks around the clock.

Despite Operator’s promising features, OpenAI is not the first to introduce this type of tool. Anthropic, a rival AI startup, released a similar feature in October 2024 called “Computer Use”.

This feature enables Claude, Anthropic’s generative AI platform, to navigate computers like a human—clicking buttons, filling out forms, and using the software.

Google is also making strides in the AI assistant race. In December, it launched Gemini 2.0, its latest generation of generative AI models.

Google’s focus is on creating AI that can perform complex tasks, like independently navigating websites, finding supplementary information, and analyzing documents.

For now, all companies emphasize that their AI tools operate under human supervision. While these systems can, for example, add items to an online shopping cart, they cannot yet complete purchases by clicking the payment button.

OpenAI’s release included a demonstration video showing Operator in action. In the example, an engineer asks Operator to find a recipe and add the required ingredients to an online shopping cart.

The AI navigates a cooking website, asks the user follow-up questions, and prompts them to log in when necessary.

Operator and similar tools represent a step closer to the vision of AI as a versatile, efficient assistant.

However, the race among OpenAI, Anthropic, Google, and others shows that the competition to lead this field is far from over.