Meta is doubling down on its commitment to artificial intelligence (AI) and plans to create one General AI (AGI)a long-term proposal aimed at integrating this technology into all aspects of daily life, as confirmed Mark Zuckerberg, CEO of the companyin a post shared in threads.
«We came to the conclusion that we need to create general intelligence to build the products we want“, emphasized the manager and repeated:Our long-term goal is to develop general intelligence, open it up responsibly, and make it widely available so that everyone can benefit from it«.
Currently, Meta has united the two teams it had previously dedicated to AI-Fundamental AI Research (FAIR) and GenAI-, the latter dedicated to generative AI; with the aim of getting closer to his new vision of creating AI. In this sense, Zuckerberg explained: This vision requires a large dedicated infrastructure, for which the company will have around 350,000 Nvidia H100s GPUs by the end of 2024, or a general computing equivalent of 600,000 H100s. «We have developed the ability to do this on a scale that may be greater than that of any other single company.“explained the CEO.
Metaverse
Regarding the plans they have for him Metaverseanother big bet from meta, right now, Zuckerberg has not provided any information about this. At this point the manager mentioned that “The company’s vision for AI and the metaverse of virtual space are interconnected«.
Zuckerberg also highlighted the role of virtual assistants and glasses, for which she “The ideal form factor for AI to see what you see and hear what you hear«. The company is continuing to work on it Call 3its AI model, which will replace Llama 2 in the future, and according to the CEO, other models are also planned to continue to be released.safe and responsible«.