Zuckerberg's AI Manifesto: Open Models Crucial for US National Security, Competition with China

Mark Zuckerberg, CEO of Meta, has laid out his vision for the future of artificial intelligence in a new manifesto, urging for open-source AI models to be developed under American leadership. Zuckerberg argues that open models are not only safer but also crucial for national security, directly challenging the strategies of competitors like Anthropic, Google, and OpenAI, who favor closed, proprietary systems. Meta itself is a major player in the AI boom, investing a staggering $145 billion in AI infrastructure this year, with plans to reach $600 billion by 2028. The company is also aggressively recruiting top AI researchers with multi-million dollar bonuses. Zuckerberg's stance supports unrestricted competition with China, stating, "Our goal should be for American open-source models to be the best worldwide. To achieve this, the hurdles that hinder the competition of American open-source models must be removed." Meta is also releasing its own open-weight AI model, "Muse Glimmer," and plans to make the parameters of other leading models public, allowing developers to modify and build upon them. This open approach contrasts sharply with the closed models offered by rivals, which can only be accessed through their platforms. Zuckerberg's views carry significant weight in Washington, where debates about AI regulation are intensifying.

“

Open models are safer.