NEWS  AI  January 19, 2024  Reading time: 2 Minute(s)

Max (RS editor)

While the tech world continues to focus on enhancing conversational AI, Meta has set its sights on a grander objective—Artificial General Intelligence (AGI). In a recent Instagram Reel, Meta CEO Mark Zuckerberg unveiled the company's commitment to developing AGI, often described as human-level AI, and making it open source.

The announcement revealed that Meta's two prominent AI research teams, FAIR (Facebook AI Research) and GenAI, have joined forces to drive the development of AGI. Zuckerberg shared insights into Meta's long-term vision through a video, emphasizing the goal of building general intelligence, ensuring responsible open sourcing, and making it widely accessible for the benefit of all.

In the video caption, Zuckerberg stated:

"Some updates on our AI efforts. Our long term vision is to build general intelligence, open source it responsibly, and make it widely available so everyone can benefit. We’re bringing our two major AI research efforts (FAIR and GenAI) closer together to support this."

Meta's current focus includes training its next-gen model, Llama 3, and establishing a massive compute infrastructure to support its future roadmap. The company aims to deploy 350,000 H100s by the end of the year, with an overall computation equivalent of almost 600,000 H100s, encompassing other GPUs. Additionally, Meta is enthusiastic about its progress in developing new AI-centric computing devices, such as the Ray Ban Meta smart glasses.

This ambitious move into AGI comes on the heels of Meta's chief scientist, Yann LeCun, expressing skepticism in December about the imminent arrival of AI superintelligence. LeCun, a prominent French computer scientist, believes that current AI systems are still decades away from achieving any semblance of sentience. He added that we are more likely to witness AI intelligence at the level of cats or dogs before achieving human-level AI intelligence.

 COVER IMAGE BY RAWPIXEL.COM ON FREEPIK