Meta Platforms CEO Mark Zuckerberg has urged for reduced U.S. restrictions on open-source AI to enhance competitiveness against Chinese counterparts. The social media company unveiled a smaller, more customizable AI model named Muse Glimmer, tailored for Mac or PC usage with a single graphics card. Zuckerberg hinted at the imminent release of larger models in a video post accompanying a 14-page essay advocating for the democratization of AI.
The push for open-weight AI, gaining momentum due to concerns over escalating AI costs and recent cybersecurity breaches involving models from Anthropic, OpenAI, and Meta, has garnered support from Zuckerberg. Open-weight models, exposing their core components for user customization, contrast with closed models kept proprietary by companies, offering a cost-effective alternative.
Recent attention on Chinese open-weight models like Kimi K3 and the defensive use of such models by Hugging Face after a cyberattack underscore the importance of open AI. Meta, among the signatories of an open-weight AI support letter, aims to catch up with Chinese startups leading in this domain with models like Qwen3.8-Max and V4-Flash.
Meta’s stock, which has experienced a 10% drop this year, showed a nearly 3% increase in premarket trading following the AI model announcement. The company’s upcoming Muse Spark 1.2 release, developed by a specialized team, signifies Meta’s commitment to open models after a lukewarm reception of Llama 4 prompted a shift in strategy.
In a comprehensive AI essay, Zuckerberg unveiled a $1 billion fund to address concerns surrounding data center expansions, crucial for AI infrastructure. He highlighted the need for policy revisions to propel U.S. firms to the forefront of open-weight models, emphasizing data utilization and distillation methods for model training efficiency. Zuckerberg also pledged to implement a governance structure to ensure model safety criteria approval by independent directors.
As the U.S. faces challenges in infrastructure development compared to China, Zuckerberg stressed the importance of minimizing regulatory hurdles for American open-source models to excel globally. The Trump administration’s decision not to subject open-weight AI models to voluntary safety tests further underscores the evolving regulatory landscape within the AI sector.