The arrival of GPT-4, the new version of ChatGPT for artificial intelligence

in spite of chat Talked about recently, the truth is that this technology was created two years ago and the tool that makes it possible, GPT-3, It was born in 2020, so after its growth a new version is expected.

During an event in Germany, engineers from Microsoft They confirmed that the update will become a reality in the third week of March with a great news package starting with higher quality in their responses.

The American company participates in gpt After your investment Open AIthe company that owns the technology, and after integrating it into the Bing browser, which now has a chatbot, this has represented an increase in the recent number of users, exceeding 100 million daily active accounts.

You may be interested: How to Create an AI Photography Studio

This technology is behind running chat To produce answers, texts and all the content that marked the direction of this artificial intelligence (AI).

However, it has limitations due to the capabilities the development engineers had at the time, such as, for example, it does not have data after 2021 and that is why some answers that include events after that date are incorrectly answered, such as I don’t know who Won the 2022 World Cup.

It may interest you: How to choose the length of ChatGPT responses

Due to its progress, the new version is necessary and it is the one that will be released soon, which is being promoted by microsoft, It is about to launch a system called Kosmos-1, which consists of a Large Multimedia Language Model (MLLM), meaning that it is an artificial intelligence that is able to receive input through text, images, audio or video clips, which gives it more possibilities to understand user requests and be More accurate results.

For this, GPT-4 It will have many more parameters than its previous version; ie with more data than is necessary to generate learning. The increase will be from 175 thousand million from the current version to 100 trillion for its update.

This will allow you to respond to more complex situations and generate more “human” responses, according to engineers at Microsoft.

It may interest you: The Application of AI to Mental Health: Lack of Controls Can Have Dire Consequences

An example of all this is that this technology will be able to receive input via video, and in this way it can learn from the displayed images, the movements and feelings of the person, the voice, the intonations of the voice, and the fluency of the conversation to extract that information and create content, which could be a summary of a movie or a conference.

Microsoft engineers gave another practical example in the world of work. They talk about the importance of the new version in areas such as customer service, where the technology can transcribe and summarize conversations, improve worker time, and save 500 man-hours a day by taking 30,000 calls a day.

“It’s not about replacing jobs, but about performing repetitive tasks in a different way than before,” said Andreas Braun, Chief Technology Officer of Microsoft Germany.

It may interest you: They thought their loved ones were asking them for help: It was an AI scam

In parallel with all this, they assure the company that artificial intelligence will continue to make mistakes and it is necessary to verify the correctness of their information. Similarly, Sam Altman, CEO of OpenAI, explained that it’s important to balance expectations with this progress, because “people are crying out to be disappointed, and that’s what will happen.”