The idea on which computers anchor has been invariably the same for more than 60 years. Perhaps for this reason OpenAIthe ChatGPT matrix, wants to light up a completely new generation of computers (and AI will play an absolutely central role in them).

To date, artificial intelligence (AI) tools such as ChatGPT allowed us to write emails, generate images and learn about specific topics on the computer (or, failing that, on the mobile phone). Sam Altman, CEO and founder of OpenAI, however, wants to go one step further and is already working on a new type of AI that goes by the name “software agent.”

This AI could partially take control of the user’s computer to independently execute complex tasks. in which human intelligence previously entered the scene.

The revolutionary new type of AI that OpenAI currently has in the oven will be able to switch between using different applications on the user’s computer and perform clicks, movements with the cursor and write text. “Agent software”, the new AI from the ChatGPT company, would be able, for example, to create tables, prepare a presentation or book a trip without the user needing to undertake any task. The user would limit himself to explaining the desired result to the AI ​​and it would take over the entire work without further mediation on its part.

“Agent software” would also have the ability to navigate the Internet to automatically compile information and provide the tools and resources necessary to carry out the task entrusted to them by the user.

If a specific graphic design program is, for example, best suited for editing an image, “agent software” would download it and then undertake the assigned task.

Google and Meta would also have artificial intelligence similar to that projected by OpenAI in the oven

Altman refers to “agent software” as a “super intelligent personal assistant” which fundamentally makes life easier for the user. «At some point you will be able to simply ask the computer for what you need and it will do the tasks for you. The benefits will be enormous,” emphasizes the CEO of OpenAI.

The company led by Sam Altman is not, in any case, the only company that currently has an AI of these characteristics in its portfolio. Google and Meta would also be working on similar projects. The idea is that the user gives up control of the computer to the AI, so that it can undertake tasks autonomously. It goes without saying that the technology projected by OpenAI would absolutely radically change the way we interact with technological devices.

However, About the new and ambitious OpenAI project there are many questions and also some concerns (which have nothing trivial). Not in vain, for the AI ​​”agent software” to function correctly, the user would have to give up control of their computer to OpenAI (since the technology would have to have access to all the applications and data stored on the device). And this inevitably sows anxiety from the point of view of privacy, security and responsibility.

Besides, OpenAI AI must be designed in such a way that it respects the user’s goals and values ​​and does not act against them, something that seems quite complicated, especially if the intelligence becomes faster and smarter than the user (in some areas at least).

OpenAI is, in any case, perfectly aware of the challenges that its project entails and is already working on solutions to make “agent software” a totally safe and reliable AI. The new ChatGPT matrix AI is currently in development and it is still unknown when exactly it will hatch to be made available to the general public.