OpenAI Launches ChatGPT Agent Capable Of Handling Complex Tasks

OpenAI introduced a new feature called ChatGPT Agent on Thursday. The new capability allows the chatbot to handle complex tasks and combine multiple features.

According to the announcement, ChatGPT Agent can access users’ computers and data and perform tasks such as sharing briefs of upcoming events, analyzing data, and creating documents on its own virtual computer.

“ChatGPT can now do work for you using its own computer, handling complex tasks from start to finish,” wrote OpenAI. “ChatGPT will intelligently navigate websites, filter results, prompt you to log in securely when needed, run code, conduct analysis, and even deliver editable slideshows and spreadsheets that summarize its findings.”

ChatGPT can now do work for you using its own computer.

Introducing ChatGPT agent—a unified agentic system combining Operator’s action-taking remote browser, deep research’s web synthesis, and ChatGPT’s conversational strengths. pic.twitter.com/7uN2Nc6nBQ

— OpenAI (@OpenAI) July 17, 2025

The new feature began to roll out for Pro, Plus, and Team customers and can be activated through the chat’s tools dropdown. The AI company promised to add more features and optimize capabilities soon.

In January, OpenAI launched Operator, a feature that allows ChatGPT to take control of the user’s computer and perform tasks autonomously. The company explained that the AI agent combines Operator’s capabilities to interact with websites and the Deep research—launched in February—to process and synthesize information, as well as ChatGPT’s conversational skills and advanced technology.

OpenAI also clarified that while ChatGPT Agent can access more information and perform more tasks, the users are “always in control.”

“ChatGPT carries out these tasks using its own virtual computer, fluidly shifting between reasoning and action to handle complex workflows from start to finish, all based on your instructions,” wrote OpenAI. “Most importantly, you’re always in control. ChatGPT requests permission before taking actions of consequence, and you can easily interrupt, take over the browser, or stop tasks at any point.”

The new feature would compete against other AI agents in the market developed by companies such as Anthropic, the Chinese Zhipu, and Microsoft.