ChatGPT gets smarter, helping users with their computer work

OpenAI, a pioneer in the AI chatbot race, has just launched ChatGPT Agent – a groundbreaking AI tool integrated directly into ChatGPT, promising to do all the work of users on the computer.

The move comes amid fierce competition from tech rivals, forcing OpenAI to constantly innovate to maintain its leadership position.

ChatGPT becomes smarter, helping users work on computers - 1

ChatGPT Agent can perform tasks on the computer for users based on their requests (Illustration: Getty).

ChatGPT Agent is built on a completely new artificial intelligence model, trained to perform many complex tasks by interacting with tools and applications on the computer.

This tool has the ability to browse text, images, send emails, and interact with users flexibly.

Users can ask ChatGPT Agent to perform a variety of tasks such as: checking work schedules to notify meetings, planning dates by finding free time and making restaurant reservations, automatically generating reports from existing data or collecting new data, creating slideshows, or sending emails on a pre-set schedule.

Some OpenAI employees have used the tool to automate daily tasks, avoiding missing or forgetting tasks.

While the response speed is slow at the moment, OpenAI says this isn’t a major issue because ChatGPT Agent will automatically execute requests in the background, allowing users to continue their work. The tool will be rolled out immediately to users who subscribe to ChatGPT Pro, Plus, and Team plans.

However, the convenience of ChatGPT Agent comes with privacy and data security concerns. The fact that this tool can access and use data and software on personal computers poses a big challenge.

OpenAI acknowledges this risk and has limited the data access that ChatGPT Agent has. Some sensitive features, such as sending emails or booking restaurant tables, will require user supervision and permission.

To ensure safety, the AI tool is also trained to reject high-risk tasks like bank transfers, to prevent hackers from taking advantage.

OpenAI CEO Sam Altman also frankly admitted: “I will explain this to my family as something advanced and experimental, an opportunity to experience the future. But this is not something I would use for important purposes or for work that involves a lot of personal information."

The development of AI virtual assistants like ChatGPT Agent reflects the general trend of the technology industry, towards creating tools that can support people in all daily tasks, moving closer to the image of virtual assistants in science fiction movies.

However, over-reliance on AI and the risk of personal data being exploited remain issues that need serious attention.

Source: https://dantri.com.vn/cong-nghe/chatgpt-tro-nen-thong-minh-hon-lam-giup-nguoi-dung-cong-viec-tren-may-tinh-20250718121619042.htm