ChatGPT has become increasingly popular ever since it was unveiled. People now use it for all sorts of everyday administrative tasks with varying degrees of success, but that does not mean that it has reached its peak or that its creators have stopped trying to improve its functions. Tat is exactly why OpenAI has just rolled out a major upgrade to ChatGPT, to continue to improve on the technology, and they have called it ChatGPT Agent. In theory, this new update should be able to perform more complex tasks, like thinking in steps, making decisions, using tools, and even fixing mistakes as they happen.
While this upgrade sounds really good, especially considering the limitations of the current AI, implementation will be where all the flaws come in and where we will be able to see the new features in the real world and not in a controlled company environment.
The new ChatGPT Agent, a feature that could solve some of the current AI’s limitations
Sam Altman, OpenAI’s CEO explained about the new tool “Agent represents a new level of capability for AI systems and can accomplish some remarkable, complex tasks for you using its own computer.” But it still has plenty of limitations that will need to be taken into account. He continues in his X post that he would “explain this to my own family as cutting edge and experimental; a chance to try the future, but not something I’d yet use for high-stakes uses or with a lot of personal information until we have a chance to study and improve it in the wild.”
With that word of caution out of the way, ChatGPT Agent can do a lot of new things, it is designed to take on multi-step projects, whether for work or personal life, which means that you can ask it to plan an event, handle research, or create a well-organized presentation from a mountain of data, but that still does not mean that the results should not be double checked. To test it, the team that improved the tech asked the AI to plan a wedding, and it did a fairly decent job of handling everything from picking an outfit to comparing flights and choosing a gift.
But where it shines continues to be in decreasing the amount of manual work that you have to do to sift through data, for example, it can pull data from websites and package it into editable document or create a spreadsheet out of data that you give it, saving you a lot of time that can be better spent elsewhere.
Since this tech is a lot more sophisticated, it will not be cheap for those who want to use it. To access these new capabilities, you will need to be on a paid ChatGPT plan, either Pro, Plus, or Team. If you already are on one of these plans, you can just head to the tools menu and turn on “Agent mode” to activate it. This will allow you to type the prompt you need and lay back while it works. Still, you will need to pay attention to see if what you are getting is what you asked for. Remember, this is still in testing and AI is not autonomous just yet.
Plus, there is always a risk of it misinterpreting a request or being misused if someone sets it up the wrong way, so never rely on it in a way that may put your personal information at risk o your job under review.
Having said that, this is a huge leap in tech and Altman could not be happier. As he explained in another X post “Watching ChatGPT Agent use a computer to do complex tasks has been a real ‘feel the agi’ moment for me; something about seeing the computer think, plan, and execute hits different.”