SAN FRANCISCO, Nov 24 — OpenAI, the company behind ChatGPT, is reportedly developing a new type of artificial intelligence, codenamed Q.* This project represents the first steps of the firm’s foray into a new generation of AI, designed to emulate human cognition as closely as possible.
According to an article published by The Information, OpenAI’s teams are working on a new model of artificial general intelligence (AGI) that could eventually be capable of “surpassing” humans in many tasks, but also potentially dangerous in the long term depending on how it’s put to use.
At the start of 2023, numerous rumours suggesting that the next major evolution of ChatGPT could be the beginning of some kind of artificial general intelligence were already circulating. The project is reportedly codenamed Q* (pronounced “Q-Star”). Eventually, such an intelligence model could equal, or even surpass, the human in various processing tasks. After becoming adept at generating texts and translation, artificial intelligence could soon excel in mathematics. For the time being, the model is only capable of solving relatively simple problems, but in time this form of intelligence could be applied to scientific research.
AGI and ASI (artificial superintelligence) remain the stuff of science fiction, as the concept suggests a reasoning capacity close to human intelligence, something which has not been approached to date. In fact, this refers to a form of artificial intelligence with the capacity to understand, learn and apply its own knowledge in the same way that a human does, and even to fill in for a human in various contexts. In its fully realized capacity, an AGI model could theoretically perform any intellectual task that a human is ordinarily capable of accomplishing. But many experts in artificial intelligence have already warned of potential pitfalls.
Even if OpenAI’s progress is impressive, creating a true AGI remains a long-term objective. There are reports, however, that the announcement of these internal advances may have had something to do with the ousting of OpenAI CEO Sam Altman by the board of directors, before his rocky return a few days later amid the threat of resignation from the majority of the start-up’s employees. — ETX Studio