Mark Zuckerberg, the CEO of Meta, has joined the race to develop Artificial General Intelligence.
While Zuckerberg has not defined a specific timeline or provided an exact description of AGI, he is committed to its creation.
In an interview, he also revealed his plans to relocate Meta’s AI research group, FAIR, closer to the team working on generative AI products within Meta’s apps, aiming to facilitate the direct application of AI breakthroughs to the company’s vast user base.
Limited engineers
Zuckerberg acknowledges the intense competition for AI talent in the tech industry, with numerous companies vying for a limited pool of researchers and engineers.
This competition has driven up compensation packages for experts in the field, with some commanding annual salaries exceeding $1 million.
CEOs like Zuckerberg are actively involved in recruitment efforts to secure top talent and prevent researchers from defecting to rival companies.
Computing power
Apart from talent, another critical resource in AI development is computing power.
Zuckerberg announced that Meta is poised to become a dominant player in this regard.
By the end of the year, Meta is set to possess over 340,000 of Nvidia’s H100 GPUs, the preferred chip for building generative AI models.
This substantial GPU inventory is expected to surpass the capacity of any other individual company, signaling Meta’s commitment to advancing AI technology.
The Definition and Evolution of AGI
The concept of AGI lacks a precise definition, and Zuckerberg, like many others in the field, does not provide a concise description.
AGI represents a form of intelligence that encompasses diverse capabilities, including reasoning, intuition, and advanced problem-solving.
According to Zuckerberg, AGI’s arrival will likely be a gradual process rather than a singular event, making it challenging to pinpoint a specific threshold.
