At an event held in London, Meta announced its imminent release of Llama 3, the latest iteration of its open source large language model designed for powering generative AI assistants. The confirmation follows a recent report by The Information hinting at Meta’s impending launch.
Nick Clegg, Meta’s president of global affairs, revealed that the rollout of Llama 3 is slated to commence within the next month, with various versions boasting diverse capabilities scheduled for release throughout the year.
Meta’s Chief Product Officer, Chris Cox, outlined plans to integrate Llama 3 across multiple products within the Meta ecosystem. The move comes as Meta endeavors to compete with OpenAI’s ChatGPT, which gained widespread popularity over a year ago.
With Llama 3 poised to offer enhanced capabilities and address previous criticisms of its predecessors’ limitations, Meta aims to broaden its appeal to users by facilitating more accurate responses and tackling a broader range of topics, including controversial issues.
While Meta remains committed to open source development with its Llama models, it adopts a cautious approach, particularly concerning other forms of generative AI such as image generation, as indicated by the decision to withhold the release of Emu, its image generation tool.
Despite Meta’s push towards Llama 3, notable skepticism towards generative AI persists within the company, with Yann LeCun, Meta’s chief AI scientist, advocating for alternative approaches like joint embedding predicting architecture (JEPA) as the future of AI.
As Meta prepares to usher in Llama 3, the debate over the future direction of AI continues within its ranks, underscoring the complexities of technological advancement and innovation.
