the new features of OpenAI for its AI – L’Express

“On November 30, we discreetly put a first version of ChatGPT online for research purposes. And it went pretty well,” joked Sam Altman, the company’s boss. “We now have around 100 million active users every week,” he added on stage in San Francisco, during a conference broadcast live online.

These millions of people use ChatGPT to write messages, ask for a cooking recipe or invent a story to tell their children – which the conversational robot can then read to them. Users subscribed to the interface will soon be able to go further, by creating their own personalized chatbot, without even needing to know how to code.

Because OpenAI launched “GPTs” on Monday, agents that “can, for example, help you learn the rules of any board game or teach mathematics to your children.”

Among the examples put online by the company, we find “Laundry Buddy”, that is to say “The Laundry Buddy”, from whom you can “ask everything about stains, machine settings washing and sorting clothes” or “The Negotiator”, which “helps you defend your interests and obtain better results”.

These chatbots are based on OpenAI language models (the core technology of generative AI) and instructions and documents provided by the chatbot creator (like the rules of a game). Widely considered a revolution comparable to the advent of the internet, generative AI makes it possible to produce texts, lines of code, images and sounds upon simple request in everyday language.

For many observers, it will notably make it possible to create these personalized agents, which will help humans in their personal and professional lives. “As (artificial) intelligence is integrated everywhere, we will all have superpowers on demand,” promised Sam Altman.

The unprecedented success of ChatGPT and the concerns raised by generative AI have propelled the young leader (38 years old) from parliamentary hearings to interviews with heads of state in 2023. On Monday, he reiterated his confidence in the ability to AI in the future to give emancipatory capacities to everyone, “on a scale that we have never seen before”. “We will be able to do more, create more and have more,” he assured.

In addition to GPTs, OpenAI’s announcements were primarily aimed at the more than 2 million developers who use its technologies to create generative AI applications. The boss unveiled “GPT-4 Turbo”, a new model that can take more context into consideration during queries, is trained on more recent data and costs less.

The application programming interface (API) gains multimedia capabilities (computer vision, voice, etc.). And OpenAI now plans to cover legal costs in the event of prosecution for intellectual property infringement, like Google or Microsoft.

For expert Yory Wurmser, the main announcement remains personalized agents. “These initiatives are similar to what Meta recently announced,” he notes, referring to the generative AIs with personalities launched this fall by the social media giant.

These new features “aim to allow OpenAI to stay ahead of companies like Google and Amazon,” according to the Insider Intelligence analyst. The world’s cloud leaders – where the generative AI training happens – Amazon, Microsoft and Google “have APIs for several major language models and will undoubtedly follow with similar tools”, he said. added.

Tech giants are engaged in a frantic race to deploy this technology, with new tools for their search engines, platforms and productivity software. The Windows manufacturer has returned to the forefront of the technological scene thanks to its major investments in OpenAI. “We love you,” said Satya Nadella, the boss of Microsoft, on Monday.

“I’m excited about the idea of ​​us building artificial general intelligence together,” replied Sam Altman, referring to OpenAI’s longer-term project: creating an AI with cognitive capabilities superior to those of humans.