Meta Boss Mark Zuckerberg Bets Big On Artificial General Intelligence To Take On OpenAI, Google. Know What AGI Is

While AGI holds the promise of revolutionary advancements, concerns loom among industry experts and AI scientists about the potential risks associated with runaway superintelligence.

Meta CEO Mark Zuckerberg shared his keen interest in the development of Artificial General Intelligence (AGI), an advanced form of artificial intelligence surpassing human intelligence in various domains. Diverging from conventional approaches, Zuckerberg has committed to responsibly open-sourcing AGI and making it accessible to the public.

Addressing the audience through a video on Threads, Meta's alternative to Twitter, Zuckerberg emphasized the necessity of AGI for creating superior AI chatbots and assistants. He stated, "The evolution of services demands the construction of full general intelligence," asserting Meta's long-term vision to open-source AGI responsibly for widespread benefit.

This strategic move by Meta follows the footsteps of industry counterparts like OpenAI and Google, all heavily investing in generative AI to enhance their core functions through advancements in Large Language Models (LLMs). Google recently introduced Gemini, the latest LLM model family, while Meta launched Llama 2 LLM, contributing to applications like the Meta AI chatbot and Ray-Ban Meta smart glasses.

Noteworthy initiatives by Meta include consolidating its AI research teams for increased synergy and establishing new infrastructure capable of accommodating around 350,000 H100 GPUs from Nvidia by the year-end.

While AGI holds the promise of revolutionary advancements, concerns loom among industry experts and AI scientists about the potential risks associated with runaway superintelligence. A reported letter from staff researchers at OpenAI to the board highlighted concerns about the threat a powerful AI discovery poses to humanity, especially before the removal of Sam Altman from the CEO position.

What Is AGI?

The concept of AGI involves the ability to devise novel solutions to unfamiliar tasks, akin to human problem-solving. Despite the aspirations of various companies, as of now, AGI remains an elusive goal in the realm of artificial intelligence.

In the perspective of Meta's rival, OpenAI, the successful creation of AGI could elevate humanity by increasing abundance, boosting the global economy, and facilitating groundbreaking scientific discoveries. However, this potential comes with substantial risks, including misuse, accidents, and societal disruptions, as cautioned by OpenAI.