According to The Wall Street Journal, Meta, the parent company of Facebook, is in the process of developing a new artificial intelligence model. The newly developed open-source AI model aims to rival OpenAI’s most powerful system.
The artificial intelligence chatbot ChatGPT released by OpenAI quickly gained a large user base shortly after its release. With ChatGPT rapidly gaining popularity on a global scale, AI technologies have once again become a focal point of interest.
As artificial intelligence projects continue to gain intense interest, many corporate companies have started investing in this field. Recent news reports have also mentioned rumors about technology giants such as Microsoft and Meta developing AI projects.
Meanwhile, as artificial intelligence projects continue to attract more attention, news broke that Meta is working on a new AI model. The Wall Street Journal stated that Meta’s new project aims to rival OpenAI.
A New Model to Compete with OpenAI’s System is Being Developed
According to a recent article shared by The Wall Street Journal, Meta, the parent company of Facebook, is in the process of developing a new artificial intelligence model. The article states that the new open-source AI model aims to rival OpenAI‘s most powerful system.
In previous months, the Financial Times also reported that Meta was preparing to release an AI chatbot similar to ChatGPT. Furthermore, Meta CEO Mark Zuckerberg, who has made numerous statements on the subject recently, expressed that the company has been working intensively on AI and Metaverse projects for many years.
Sources from The Wall Street Journal stated that Meta’s plans for the new system are currently open-source, allowing other companies to create AI tools for producing high-level texts, analysis, and other output types. As the company builds the necessary data centers to create such a high-level system, it has started purchasing more of Nvidia’s H100 semiconductors, one of the most powerful chips currently available on the market. Sources said that Meta expects training for the large language model (LLM) to begin in early 2024 and be ready for release next year. It was also mentioned that it is likely to be released after Google’s upcoming LLM Gemini.
Microsoft, one of OpenAI’s main supporters, has collaborated with Meta to help make Llama 2 available on its Azure cloud computing platform. However, sources stated that Meta plans to train its upcoming model on its own infrastructure.
