ChatGPT Agent Mode: A New Era of Automation and Research
ChatGPT's new Agent Mode is an exciting development that merges the capabilities of their existing operator and deep research products. For those unfamiliar, the operator function allows the AI to take control of a web browser to complete tasks, while deep research involves conducting extensive web searches and providing detailed reports. Agent Mode combines these functionalities, offering a more comprehensive toolset.
In addition to these features, Agent Mode can perform tasks such as inputting data into spreadsheets, creating presentations, running code, and accessing a file system. These added capabilities provide users with a versatile tool that can handle a variety of tasks efficiently.
During my initial experience with Agent Mode, I was able to run two prompts. The first prompt involved the AI working for about half an hour to organize data into a spreadsheet, which it executed successfully. Observing the process and the final output was quite satisfying. However, I found the interface for editing the spreadsheet output somewhat cumbersome and in need of improvement. It would be beneficial if the spreadsheet could be edited on a more intuitive platform.
Unfortunately, my follow-up prompt did not complete, likely due to the ongoing rollout of the feature. Despite this, the potential of Agent Mode is evident. It represents the next evolution of the deep research and operator products, offering a richer and more thorough experience by integrating web browsing, file system access, and code execution.
As the rollout progresses, users with pro accounts can look forward to receiving 40 queries a month. This will undoubtedly lead to innovative uses and applications of the tool. I am eager to see how people will leverage these capabilities to create new and interesting solutions.