Photo by energepic.com on Pexels

OpenAI is pushing the boundaries of AI assistance with the launch of ChatGPT Agent, a powerful tool granting the chatbot complete control over a ‘virtual computer’ to execute complex tasks autonomously. This new model, specifically engineered for the Agent, is trained on intricate operations demanding the integration of various tools including a text browser, visual browser, and terminal.

Imagine ChatGPT Agent managing your calendar, planning your meals, crafting presentations, and handling other tedious tasks. This is made possible by combining functionalities from OpenAI’s previous Operator and Deep Research tools into a single, potent agent.

Early demonstrations showcase the Agent’s capabilities, ranging from orchestrating date nights to generating in-depth research reports. Importantly, the system is designed with user control in mind, seeking permission before executing irreversible actions like sending emails or making reservations.

Safety measures are paramount. OpenAI has implemented safeguards to prevent misuse concerning ‘high biological and chemical capabilities’ and has restricted financial transactions. The rollout will begin with Pro, Plus, and Team users, followed by ChatGPT Enterprise and Education subscribers. Unfortunately, users in the European Economic Area and Switzerland will have to wait, as a specific timeline for their access has not yet been announced.