GPT-4o, OpenAI’s newest AI model, is revealed.

The most recent artificial intelligence big language model was introduced by OpenAI on Monday. According to the company, this model would make ChatGPT more intelligent and simpler to manage.

In comparison to the GPT-4 model that was introduced by the firm a little more than a year ago, the new model, which is referred to as the GPT-4o, is an improvement.

The model will be made accessible to consumers who do not pay for it, which means that everyone will have access to the most cutting-edge technology being offered by OpenAI via ChatGPT.

According to the demonstration that the business gave on Monday, GPT-4o would successfully transform ChatGPT into a digital personal assistant that is capable of engaging in spoken discussions in real time.

Moreover, it will be able to communicate with people via the use of language and “vision,” which means that it will be able to see screenshots, images, papers, or charts that have been submitted by users and engage in a discourse about them.

It was said by Mira Murati, Chief Technology Officer of OpenAI, that the new version of ChatGPT would now also have memory capabilities. This means that it will be able to learn from prior discussions with users and will also be able to do translation in real time.

During the live demonstration that took place from the headquarters of the firm in San Francisco, Murati said, “This is the first time that we are really making a huge step forward when it comes to the ease of use.” “This interaction becomes a great deal more natural and a significantly simpler process.”

OpenAI strives to keep ahead of the expanding rivalry in the artificial intelligence arms race, and the latest release arrives at the same time.

There have been efforts made by competitors like as Google and Meta to develop more powerful big language models. These models are used to power chatbots and can also be used to apply artificial intelligence technology to a variety of other goods.

The OpenAI event took place one day before Google’s annual I/O developer conference, which is anticipated to be the occasion at which the company would disclose that it has updated its Gemini AI model.

Google’s Gemini is multimodal, which means that it can comprehend and create text, graphics, and audio. This is similar to the new GPT-4o which is also multimodal.

The update from OpenAI also comes ahead of the anticipated announcements from Apple on artificial intelligence that will take place at the Worldwide Developers Conference the following month.

These announcements may include new methods to incorporate AI into the next iPhone or iOS updates.

As for Microsoft, which has spent billions of dollars into OpenAI in order to incorporate its artificial intelligence technology into Microsoft’s own products, the most recent release of GPT may prove to be a benefit for the company.

OpenAI executives performed a spoken discussion using ChatGPT in order to get real-time instructions for solving a mathematical issue, to tell a narrative to their children before sleep, and to receive advise on how to code. The ChatGPT was able to talk in a voice that sounded genuine and human, as well as a voice that sounded like a robot, and it even sung a portion of one of the responses. One other thing that the gadget was able to do was examine a picture of a chart and talk about it.

In addition, they demonstrated that the model was able to identify the feelings of users; for example, it listened to the breathing of an executive and provided him with encouragement to calm down.

The female voice of ChatGPT, which sounds eerily similar to the digital companion spoken by Scarlett Johansson in the film “Her” from 2013, made a joke about the staff member by saying, “You’re not a vacuum cleaner!”

Additionally, ChatGPT was able to carry on a discussion in a number of other languages by automatically translating and replying to the chat. According to OpenAI, the tool is now capable of supporting more than fifty different languages.

“The new voice (and video) mode is the best computer interface I’ve ever used,” stated OpenAI CEO Sam Altman in a blog post that was published after the announcement was made. “It has the sense of artificial intelligence from a movie, and the fact that it is genuine is still a little bit strange to me.

Arriving to reaction speeds and expressiveness levels comparable to those of humans turns out to be a significant shift. Using the GPT-4o capabilities, OpenAI will create a ChatGPT desktop software, according to Murati.

This will provide consumers with an additional platform on which they may engage with the technology that the business has developed. Developers who are interested in building their own unique chatbots will now have access to GPT-4o via OpenAI’s GPT shop.

This functionality, which was previously only available to paying users, will now also be accessible to non-paying users. Within the next several months, ChatGPT will be receiving the new technologies and features that have been developed.

Free ChatGPT users will only be able to interact with the new GPT-4o model for a limited number of times before the tool will automatically return to using the older GPT-3.5 model. Paid ChatGPT subscribers, on the other hand, will have access to a bigger number of messages using the most recent model.

ChatGPT is already being used by more than one hundred million users, according to OpenAI.

It is possible that an upgraded ChatGPT experience, as well as the capability to connect with it on desktop environments and via enhanced audio interactions, might provide even more individuals with a cause to utilize its technology.

The actions are being taken at a time when Google and Meta are incorporating artificial intelligence into consumer products that are more commonly utilized, such as Instagram and Google Assistant. This may make the technology of those firms more accessible to a wider audience and more frequently used.