Alphabet, the parent company of the search engine, is concerned about the dazzling success of ChatGPT, Dalle-E and Stable Diffusion. Three tools with stunning performance, all based on artificial intelligence. Explanations.
General alert at Google. According to several specialist media, all of which cite an article by New York Times published Wednesday, December 21, the tech giant has just triggered an internal a “red code”. Translation: it’s panic and we have to react. Alphabet, the parent company of the search engine, is concerned about the dazzling success of ChatGPT, Dalle-E and Stable Diffusion. Three tools with stunning performance, all based on artificial intelligence.
The first, developed by the OpenAI organization, one of whose presidents is none other than Elon Musk, is a conversational robot, capable of understanding and answering questions asked in natural language, explains 01Net. “Powered by OpenAI’s GPT-3 algorithm, ChatGPT impresses with its speed of execution and its ability to provide answers which, although sometimes quite mechanical, are most of the time, quite relevant, explain our colleagues . The robot can answer just about all your queries.”
A painting of a fox sitting in a field at sunrise in the style of Claude Monet
It is possible to test its performance, simply by registering on the site (with a Google account in particular…). And it’s true that the result impresses with its speed and relevance. Unheard of until now even if obviously everything is not yet perfect. Examples.
We ask him a simple question: “What is your name?” Here is his response, generated in a handful of seconds: “My name is Assistant. I am a large scale language model trained by OpenAI. I’m designed to help people find answers to their questions and solve problems by providing clear and precise explanations. I don’t have a proper name like a real person, but I’m happy to help you any way I can.”
Another example: “Who is Georges Frêche?” ChatGPT, almost immediately: “Georges Frêche was a French politician, member of the Socialist Party. He was elected deputy for the Montpellier constituency in 1981 and was re-elected several times until 2007. He was also mayor of Montpellier from 1977 to 2008. Frêche was known for his controversial positions and his provocative remarks, which earned several convictions for incitement to racial hatred and discrimination. He passed away in 2010.”
Dalle-E is also a project of the OpenAI organization, totally different but equally impressive. It allows images to be generated from a description, always in natural language. Like: “A painting of a fox sitting in a field at sunrise in the style of Claude Monet.” Stable Diffusion works on the same principle.
The beginnings of a technological upheaval that Google cannot ignore. According to Digital, Sundar Pichai, the company’s managing director, “has held a series of meetings aimed at redefining the firm’s priorities in this area. The work of many groups has been disrupted to respond to the threat posed by ChatGPT, according to a memo and audio recording.
01Net recalls that until now, Google has been cautious about this type of tool, in particular because there is no question for a group with such importance and such a reputation of opening up a potentially as impactful service to the general public without that it is perfectly in focus. “Google is indeed afraid of unleashing a monster on the web that would provide false information or hate speech, explains the site specializing in new technologies. Both Microsoft and Meta have already paid the price for being too eager to launch conversational AI.