Friday, 5 September 2025
26.7 C
Singapore
28.3 C
Thailand
19.4 C
Indonesia
28.3 C
Philippines

OpenAI’s new ChatGPT Agent can now complete real-world tasks for you

ChatGPT Agent from OpenAI helps you book, plan, and generate reports with ease, bringing AI task completion into your daily life.

You can now do more with ChatGPT than ask questions or generate text. OpenAI has introduced a brand-new feature called ChatGPT Agent. It’s designed to help you complete actual tasks, such as booking a restaurant, ordering ingredients, generating reports, or planning a night out. This new tool goes beyond standard AI chat functions, bringing smart assistance into your everyday life.

OpenAI created this feature by combining two of its earlier internal tools—Operator, which could navigate websites and click around like a human, and Deep Research, which could collect and summarise information into structured reports. By merging the two, OpenAI has built an intelligent assistant that not only understands what you ask but also acts on it in practical ways.

Whether you’re planning a dinner, managing your schedule, or needing a detailed report on a business trend, ChatGPT Agent can handle it all, right inside your ChatGPT interface.

A look at how ChatGPT Agent works

Let’s say you want to plan a date night. The ChatGPT Agent can check your Google Calendar to see when you’re free and then head to OpenTable to search for a restaurant and make a booking. In another demo, it was asked to generate a report on the rising popularity of a company called Labubus. It handled the request smoothly, researching, analysing and presenting the results in a slide deck format.

You don’t need to worry about the tool taking control of your device. The agent doesn’t operate on your actual computer. Instead, it works within its secure virtual environment. This setup ensures both privacy and safety, while still giving you the benefits of having a digital helper that feels hands-on.

Safety measures and availability

Because this tool is so powerful, OpenAI has implemented robust safeguards. One key restriction is that ChatGPT Agent cannot be used to create dangerous materials, such as biological or chemical weapons. Another limitation is that its memory feature has been turned off. This decision was made to prevent bad actors from using advanced prompt injection techniques to steal sensitive data.

OpenAI is aware of the risks and is taking precautions as the technology rolls out.

ChatGPT Agent is now available to Pro, Plus, and Team users. If you’re one of them, you can try it out by selecting “agent mode” from the tools menu or typing “/agent” in the chat box. It’s a simple way to access something that feels futuristic, but is very real.

While it’s still early days, ChatGPT Agent represents a major shift in how artificial intelligence can support your daily tasks. It’s no longer just about having a conversation—it’s about getting things done.

Hot this week

Veeam launches first software appliance for instant, secure data protection

Veeam has launched its first hardware-agnostic software appliance, offering instant, secure data protection with built-in resilience.

TikTok introduces voice messages and image sharing in direct messages

TikTok is rolling out voice messages and image sharing in direct messages for all eligible users, boosting its social features.

Shoppers face conflicting advice from ChatGPT and Google’s AI tools

A study reveals that ChatGPT and Google AI frequently disagree on brand recommendations, with notable differences in transparency and citation levels.

Airwallex acquires OpenPay to expand billing and subscription capabilities

Airwallex acquires OpenPay to add billing and subscription management tools, aiming to create the first truly global billing platform.

AMD could challenge NVIDIA with the next-generation flagship GPU

AMD is rumoured to be developing UDNA GPUs for a 2027 launch, aiming to rival NVIDIA’s flagship models with a unified architecture.

HubSpot unveils Loop Marketing playbook to drive growth in AI era

HubSpot launches Loop Marketing playbook and over 200 AI updates to help businesses grow in the era of AI search and zero-click results.

One in three Australian workers expose company data to AI platforms, Josys warns

Over a third of Australian workers upload sensitive data to AI tools, with Josys warning of rising risks from shadow AI and weak governance.

Singapore Polytechnic partners ESGpedia to strengthen sustainability efforts for local businesses

Singapore Polytechnic and ESGpedia partner to help Singapore businesses cut emissions, boost energy efficiency, and support the Green Plan 2030.

Veeam launches first software appliance for instant, secure data protection

Veeam has launched its first hardware-agnostic software appliance, offering instant, secure data protection with built-in resilience.

Related Articles

Popular Categories