GPT Chat with a new model more advanced with audio and visual capabilities

 OpenAI has developed a new version of GPT Chat that allows users to have smooth verbal conversations with the program, in another step towards the development of advanced assistant software through artificial intelligence technology, currently the main driver of giant technology companies.

Thanks to the new model, called GPT-4 or GPT-4o, GPT Chat will be able to understand text, sound and images, and respond by writing, speaking or generating images.

These new features will be gradually added to GPT Chat, first in text and images for subscribers to the paid service, and also for users of the free version with limited use. The new version with "Voice Mode" is expected to be released in the coming weeks for subscribers of the paid service.

This feature allows the reproduction of conversations between people in a remarkable way.

In a live video presentation, GPT Chat was able to use the smartphone camera to read the emotions on users' faces, guide them through breathing exercises, read them a story, and help them solve a math problem. Most importantly, users could easily stop the program at any time.

The software addressed an engineer at Open AI by saying, "You look happy (...) Would you like to tell me the reason for this positive mood?" The engineer replied that he was telling the audience how "useful and wonderful" it was, and the software replied, "Stop, you make me blush."

In late 2022, with the introduction of ChatGPT, which generates content based on a simple request in everyday language, OpenAI started a revolution in generative artificial intelligence that surprised all the tech giants.

Since then, Silicon Valley tech giants have been racing to develop AI tools and software that are more powerful than ever before. Google is set to unveil its latest innovations on Tuesday, while Microsoft, the main investor in OpenAI, is planning an event for journalists and developers next week.

OpenAI president Sam Altman on Friday denied rumors that the company was preparing ads. "No GPT-5+, no search engine," he announced at the event.

Altman had previously revealed his admiration for the science fiction film "Her," which tells the story of a man who falls in love with artificial intelligence by speaking to it verbally.

"It was an amazing prophecy," he said at a conference last September. "It inspired us in more ways than one...especially that we all have a personal agent trying to help us."

ChatGPT still has a long way to go before it becomes an artificial intelligence agent with deep knowledge, proactive capabilities, and the ability to adapt to users' personal desires, as the company promises. But this update has sparked mixed feelings of admiration and concern among industry experts.

"I was amazed at the extent to which the presentations succeeded in anthropomorphizing the models" to make them resemble humans, Jeff Beaudier of Hugging Face told AFP, adding, "This creates confusion and false expectations."

"People can project qualities onto models and become emotionally attached to them," explained the product manager of this collaborative and open generative AI platform. "They won't understand why models can provide false information, and they won't know in which situations to trust them or not."

Sam Altman regularly promotes his vision of an artificial intelligence that will one day become "general," that is, endowed with human cognitive abilities and capable of achieving scientific breakthroughs in the service of humanity.

OpenAI, which was originally founded as a non-profit research lab, was valued at about $80 billion when it went public last February, according to the New York Times. Since December 2023, its annual revenue has reached about $2 billion, according to the Financial Times.

"A very important part of our mission is to make all of our advanced AI tools freely available to the public (so) people can intuitively understand what the technology can do," Mira Moratti, the California-based startup's director of technology, said during the presentation.

"This is the first time we have taken a big step forward in terms of usability," she added, "This is very important. This is the future of how we interact with machines."