OpenAI unveils new AI agent said to think and act independently

OpenAI on Friday launched its new AI product ­– "ChatGPT Agent," marking the first-ever combination of ChatGPT with "Deep Research" and "Operator," making its move into the field of AI agents.

This new product boasts the ability to think and act autonomously, combining aspects of its previous agentic features, Operator, which can interact with websites, Deep Research that can conduct multi-step research for advanced tasks, and ChatGPT that can generate text. It can carry out complex, multi-step tasks, including online shopping, restaurant reservations, writing research reports, creating presentations, and performing financial analysis.

With just a single command, users can have ChatGPT Agent browse the web, click through options, filter information, run code, and generate slides or spreadsheets, significantly saving time and manpower. In other words, users only need to provide a task prompt, and the rest of the work is handled by the ChatGPT Agent, which delivers the final output.

OpenAI noted that ChatGPT Agent is especially suitable for tasks like basic financial analysis, which can reduce hours of work to under 30 minutes. This feature is now available for users of ChatGPT's Pro, Plus and Team tiers, with enterprise and education users set to gain access later this summer.

(With input from Reuters)