Zuckerberg named the main threat of artificial intelligence

**Zuckerberg reveals the main danger of artificial intelligence**

Meta Corporation founder Mark Zuckerberg has published a manifesto outlining his views on the development of artificial intelligence (AI), Axios reported. He said that the greatest danger to society from the development of artificial intelligence is the acquisition of excessive control over the technology by one organization or country.

Zuckerberg notes in his manifesto that many of the threats associated with AI are exaggerated and the positive potential of this technology is underestimated. He also noted that he supports companies working with governments in the early stages of model development, unlike the pre-release regulation that developers have often faced in recent times.

Zuckerberg published his manifesto on the same day that Meta presented its open-source AI model called Muse Glimmer. In such models, various parameters and data are publicly available, and users can download them. Although Muse Glimmer is much weaker than the most advanced models of recent times, it allows you to run SI systems directly on user devices.

Zuckerberg once again advocated open-source models, as well as the possibility of so-called “distillation” - that is, training SI models on the basis of other, more powerful models and “extracting” their key features using special queries. He believes that it is important to create conditions for everyone to use the technology. Zuckerberg also announced plans to create a $ 1 billion fund to support communities in areas near large data centers, the construction of which has accelerated due to the development of artificial intelligence.