OpenAI’s ChatGPT has taken the next step in its goals to utmost AI automation. The company has released the ChatGPT Agent, that is designed to autonomously execute a wide variety of complex computer-based tasks on behalf of users. Think of it as your personal assistant that can take care of your boring computer tasks while you are away.
The ChatGPT Agent integrates the power of OpenAI’s previously experimental systems – Operator, which was a web browser and automation tool, and Deep Research, which is an in-depth analysis tool. The combination of these tools allows the agent to not only process information but also take actions on the user’s behalf. The agent can navigate to various websites, manage calendar appointments, generate editable presentations as well as spreadsheets, and even execute code within a self-contained virtual environment.
ChatGPT Agent to take care of your computer work
“Agent represents a new level of capability for AI systems and can accomplish some remarkable, complex tasks for you using its own computer. It combines the spirit of Deep Research and Operator, but is more powerful than that may sound—it can think for a long time, use some tools, think some more, take some actions, think some more, etc,” wrote Sam Altman in a post on X.
“For example, we showed a demo in our launch of preparing for a friend’s wedding: buying an outfit, booking travel, choosing a gift, etc. We also showed an example of analysing data and creating a presentation for work,” he added.
OpenAI reveals that this new agent is powered by a dedicated AI model specifically trained for complex, multi-step agentic tasks that utilises reinforcement learning. A key feature of the ChatGPT Agent is its ability to think and iterate on plans.
Users will have control of ChatGPT Agent
OpenAI says that the agent is designed to be highly collaborative, allowing users to interrupt or steer the agent mid-task to clarify instructions or adjust its approach, all without losing progress.
“Likewise, ChatGPT itself may proactively seek additional details from you when needed to ensure the task remains aligned with your goals. If a task takes longer than anticipated or feels stuck, you can pause it, ask it for a progress summary, or stop it entirely and receive partial results. If you have the ChatGPT app on your phone, it will send you a notification when it’s done with your task,” says OpenAI in its blog post.
“We have built a lot of safeguards and warnings into it, and broader mitigations than we’ve ever developed before from robust training to system safeguards to user controls, but we can’t anticipate everything. In the spirit of iterative deployment, we are going to warn users heavily and give users freedom to take actions carefully if they want to,” says Sam Altman.
“We recommend giving agents the minimum access required to complete a task to reduce privacy and security risks,” he adds.
How to access ChatGPT Agent?
At the moment, the ChatGPT agent is rolling out to Pro, Team, and Plus subscribers. ChatGPT Pro subscribers will receive 400 queries per month whereas Team/Plus users will get only 40 queries per month.
