Meta Platforms is in the process of developing a new artificial intelligence system with the goal of matching the capabilities of OpenAI’s most advanced model, as reported by the Wall Street Journal, citing sources familiar with the situation.
The parent company of Facebook, Meta, aims to have this new AI model ready for deployment by the following year, according to the Journal’s report. It is anticipated that this upcoming system will outperform Meta’s current commercial model, Llama 2.
Llama 2, which Meta introduced as an open-source AI language model in July, has been made accessible through Microsoft’s Azure cloud services, positioning itself in competition with OpenAI’s ChatGPT and Google’s Bard.
While the specific details of the planned AI system may undergo changes, its primary purpose is to assist other companies in developing services capable of generating sophisticated text, conducting advanced analyses, and producing other valuable outputs, as detailed by the newspaper report.
Meta’s schedule for this project suggests that they intend to commence the training of this new large language model in early 2024, as per the report. At the time of the report, Meta had not yet responded to a request for comment from Reuters.
In the emerging field of generative AI, businesses and enterprises have been increasingly adopting new capabilities and improving their operational processes, especially since the launch of OpenAI’s ChatGPT at the end of the previous year.
In a related development reported by Bloomberg News in July, Apple is also actively involved in creating AI offerings similar to OpenAI’s ChatGPT and Google’s Bard. It was revealed that Apple has developed its framework, known as ‘Ajax’, for building large language models and is presently testing a chatbot that some of its engineers informally refer to as ‘Apple GPT’.